System

Abstract

A system is provided.SOLUTION: The system includes a terminal means for inputting business contents, a server means for receiving the input business contents, a natural language processing means for specifying related laws and potential risks from the received business contents, a generation AI model means for generating countermeasures corresponding to the specified risks, a document automatic generation means for automatically generating necessary documents and mails on the basis of the generated countermeasures, and an automatic transmission means for automatically transmitting the generated documents and mails to persons concerned.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Traditionally, legal responses at companies have relied on legal department staff with specialized knowledge, and because not all employees are familiar with legal matters, there has been the problem of time and effort required to check relevant laws and identify risks. Furthermore, failure to take appropriate legal action increases the risk of security and copyright issues. Furthermore, these response tasks are often performed manually, resulting in reduced operational efficiency. [Means for solving the problem]

[0005] To solve the above-mentioned problems, the present invention provides the following means. The system includes a terminal means for inputting business content, a server means for receiving the input business content, a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, a generative AI model means for generating countermeasures for the identified risks, an automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, and an automatic sending means for automatically sending the generated documents and emails to relevant parties. The system further includes a template selection means for selecting document and email templates based on the identified risks and countermeasures, enabling the generation of appropriate documents specialized for each business. Furthermore, the system includes a feedback input means for allowing users to input feedback on the content of the automatically generated documents and emails, which is reflected in improving the accuracy of the generative AI model, enabling the system's accuracy to be continuously improved. This allows even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, thereby achieving business efficiency and risk management.

[0006] The "terminal means" is a device for a user to input business details, and is a device that has the function of inputting information in text format and transmitting it to a server.

[0007] The "server means" is a central processing unit that receives and analyzes the business content sent from the terminal, and is a device that has a hub function for identifying relevant laws and regulations and risks and generating countermeasures.

[0008] "Natural language processing means" is a technology for analyzing business content received within the server and identifying relevant laws and regulations and potential risks.

[0009] A "generative AI model means" is an artificial intelligence model used to generate responses to risks identified by the natural language processing means.

[0010] The "automatic document generation means" is a device that has the function of automatically creating necessary documents and emails based on the solutions obtained from the generation AI model means.

[0011] "Automatic sending means" is a device that has the function of automatically sending generated documents and emails to related parties.

[0012] The "template selection means" is a device having a function for selecting an appropriate document or email template based on the business content and the method of dealing with the business.

[0013] A "feedback input means" is a device that has the function of allowing users to input feedback on the contents of generated documents or emails, thereby contributing to improving the accuracy of the generative AI model.

[0014] "Related parties" are people or departments that receive documents and emails related to business operations, and include users, related internal departments, external legal advisors, and the like. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[0037] System Overview

[0038] The system mainly consists of the following elements:

[0039] Terminal means for inputting business details

[0040] Server means for receiving input business details

[0041] Natural language processing means to identify relevant laws and potential risks from received business content

[0042] A generative AI model that generates countermeasures to address identified risks

[0043] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0044] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0045] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[0046] Example of a system

[0047] Terminal means

[0048] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0049] Server Means

[0050] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0051] Natural language processing and generative AI modeling tools

[0052] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0053] Automatic document generation method

[0054] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0055] Automatic transmission method

[0056] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0057] Feedback Input Method

[0058] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0059] Specific examples

[0060] For example, if a user enters "Plan a marketing strategy for new product A," the system will automatically go through the following steps to identify relevant laws and regulations, assess risks, present countermeasures, and generate and send documents.

[0061] 1. The user enters the details of the job into the terminal and sends it to the server.

[0062] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[0063] 3. The server uses a generative AI model to generate countermeasures for the identified risks.

[0064] 4. The server automatically generates appropriate documents and emails based on the corrective action.

[0065] 5. The server automatically sends the generated document to the marketing and legal departments.

[0066] 6. The user reviews the document on their device and provides feedback if necessary.

[0067] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[0068] The processing flow will be explained below.

[0069] Step 1:

[0070] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[0071] Step 2:

[0072] The device sends the entered business details to the server, and the data is securely transferred using an API.

[0073] Step 3:

[0074] The server receives the submitted business content and passes the received data to a natural language processing (NLP) engine for analysis.

[0075] Step 4:

[0076] The server uses a natural language processing engine to analyze the received business content, understand the structure and meaning of the sentences, and identify relevant laws and regulations and potential risks.

[0077] Step 5:

[0078] The server identifies relevant laws and regulations and potential risks. For example, it identifies "consumer protection laws" and "regulations on advertising display" and recognizes the risk of violations.

[0079] Step 6:

[0080] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[0081] Step 7:

[0082] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[0083] Step 8:

[0084] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[0085] Step 9:

[0086] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[0087] Step 10:

[0088] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[0089] Step 11:

[0090] The user checks the contents of documents generated on the device and emails sent, and confirms that the contents are appropriate.

[0091] Step 12:

[0092] Users can input feedback as needed, which is sent from the device to the server and used to improve the accuracy of the generative AI model.

[0093] In this way, the system automates everything from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, and improving accuracy through feedback. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, resulting in improved work efficiency and risk management.

[0094] Example 1

[0095] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0096] In the past, it was difficult for employees with insufficient legal knowledge to respond quickly and appropriately to legal issues. Furthermore, the entire process of identifying relevant laws and potential risks, developing countermeasures, and creating and sending documents required a great deal of time and effort. Furthermore, feedback on the content of generated documents and emails was not properly reflected, making it difficult to improve the accuracy of the system. This resulted in reduced work efficiency and inadequate risk management.

[0097] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0098] In this invention, the server includes a means for receiving business content, a means for identifying relevant laws and regulations and potential risks from the received business content, and a generative AI model means for generating countermeasures to address the identified risks. This allows even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters. Furthermore, by including a means for automatically generating necessary documents and emails based on the generated countermeasures, a means for automatically sending the generated documents and emails to relevant parties, and a means for users to input feedback on the content of the automatically generated documents and emails and reflect this feedback in improving the accuracy of the generative AI model, it is possible to further improve business efficiency and the accuracy of risk management.

[0099] "Business content" refers to text information related to a business that a user inputs into a terminal.

[0100] "Terminal means" refers to a device or its interface that allows a user to input business details.

[0101] The "server means" is a server that receives the business contents sent from the terminal and performs the processing.

[0102] "Natural language processing means" refers to technologies and algorithms used to analyze received business content and identify relevant laws and regulations and potential risks.

[0103] "Generative AI model means" refers to an artificial intelligence model for generating countermeasures to identified risks.

[0104] The "automatic document generation means" is a system or program for automatically creating the necessary documents or emails based on the generated solutions.

[0105] "Automatic sending means" refers to a system or method for automatically sending generated documents or emails to relevant parties.

[0106] "Feedback input means" refers to a system or method that allows users to input feedback on the contents of automatically generated documents or emails, and uses that feedback to improve the accuracy of the generative AI model.

[0107] "Template selector" means a system or algorithm for selecting document or email templates based on identified risks and countermeasures.

[0108] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[0109] System Configuration

[0110] The system includes the following major hardware and software elements:

[0111] Terminal means: A device for inputting work details. For example, a PC or tablet. The user inputs the work details and sends them to the server.

[0112] Server means: A server that receives and processes the business content sent from the terminal. Specifically, it runs natural language processing (NLP) libraries (e.g., SpaCy, BERT) and generative AI models (e.g., GPT-3).

[0113] Natural language processing means: A program that runs on the server and analyzes the received business content to identify relevant laws and regulations and potential risks.

[0114] Generative AI model means: An artificial intelligence model integrated into the server generates countermeasures for identified risks.

[0115] Automatic document generation: Automatically create documents and emails based on the solutions generated on the server. A template engine (e.g., Jinja2) is used.

[0116] Automatic sending means: A system that sends generated documents and emails to the relevant parties, for example, via an SMTP server.

[0117] Feedback input method: The user checks the contents of the generated documents and emails, and inputs feedback as necessary to improve the accuracy of the generative AI model. The feedback data is stored in a database (e.g., PostgreSQL, MySQL).

[0118] Template selection tools: Tools for selecting document and email templates based on identified risks and treatments.

[0119] Example of a system

[0120] Terminal means

[0121] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0122] Server Means

[0123] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0124] Natural language processing and generative AI modeling tools

[0125] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0126] Automatic document generation method

[0127] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0128] Automatic transmission method

[0129] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0130] Feedback Input Method

[0131] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0132] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[0133] Specific examples

[0134] For example, if a user inputs "Plan a marketing strategy for new product A," the process will go through the following steps:

[0135] 1. The business details entered by the user are sent to the server from the terminal.

[0136] 2. The server receives the business details and performs natural language processing to identify relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display."

[0137] 3. The server uses the generative AI model to generate a solution, such as suggesting "create advertising copy that complies with the Consumer Protection Act."

[0138] 4. The server automatically generates documents and emails based on the measures taken, such as a "marketing plan" or a "compliance checklist."

[0139] 5. The server automatically sends the generated document to the relevant parties.

[0140] 6. The user reviews the document and provides feedback if necessary.

[0141] Examples of prompts

[0142] An example of a prompt to input to a generative AI model is: "Analyze the laws and potential risks related to the marketing strategy for new product A, and propose appropriate countermeasures."

[0143] Based on this prompt, the generative AI model suggests appropriate actions, and the system then generates the necessary documents and emails.

[0144] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0145] Step 1:

[0146] The user inputs the details of the job into the terminal and sends it to the server.

[0147] Input: Enter the job description in text format as "Plan a marketing strategy for new product A."

[0148] Output: Text data of the work content sent from the terminal to the server.

[0149] Specific operation: The user enters "Plan a marketing strategy for new product A" into the input form on the terminal and clicks the submit button.

[0150] Step 2:

[0151] The server receives the business content and analyzes it using natural language processing means.

[0152] Input: Text data of the work content sent from the terminal.

[0153] Output: Data on relevant laws and regulations and potential risks extracted from business operations.

[0154] Specific operation: The server analyzes the text received from the device, "Plan a marketing strategy for new product A," using a natural language processing tool (such as SpaCy or BERT), and identifies relevant laws and regulations such as "consumer protection laws" and "regulations on advertising display," as well as potential risks.

[0155] Step 3:

[0156] The server uses the generative AI model to generate countermeasures for the identified risks.

[0157] Input: Data on relevant laws and regulations and potential risks identified through natural language processing.

[0158] Output: Text data of relevant laws and regulations and countermeasures for potential risks.

[0159] Specific operation: When a risk related to the Consumer Protection Act is identified, the server sends a prompt to the generative AI model (e.g., GPT-3) saying, "Please create advertising text that complies with the Consumer Protection Act," and generates a solution (appropriate advertising text and display method).

[0160] Step 4:

[0161] The server automatically generates the necessary documents and emails based on the generated solutions.

[0162] Input: Text data of solutions obtained from the generative AI model.

[0163] Output: Auto-generated documents and email data.

[0164] Specific operation: The server receives the generated solutions in text format and uses a template engine (e.g., Jinja2) to automatically generate a "marketing plan in accordance with consumer protection laws" and a "compliance checklist."

[0165] Step 5:

[0166] The server automatically sends generated documents and emails to the relevant parties.

[0167] Input: Data from automatically generated documents and emails.

[0168] Output: The sent document or email reaches the relevant person.

[0169] Specific operation: The server sends the generated "marketing plan" via email to the user, marketing department, and legal department via the SMTP server.

[0170] Step 6:

[0171] The user checks the generated documents and emails and provides feedback.

[0172] Input: The contents of the document or email sent.

[0173] Output: Feedback data from users.

[0174] Specific actions: The user reviews the received "Marketing Plan" and submits feedback by entering a comment such as "This ad copy is appropriate, but I would like it to be a little more specific."

[0175] Step 7:

[0176] The server receives user feedback and uses it to improve the accuracy of the generative AI model.

[0177] Input: Feedback data submitted by the user.

[0178] Output: Updated generative AI model data.

[0179] Specific operation: The server stores the received feedback in a database and uses the feedback data as training data for the generative AI model to improve the accuracy of the model.

[0180] (Application example 1)

[0181] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0182] Conventional regulatory compliance and risk management systems have had difficulty identifying work content and providing appropriate countermeasures quickly and accurately. In particular, with regard to work robots in factories, immediate responses to complex regulations and risks are required, but if on-site personnel lack sufficient legal knowledge, responses may be delayed. Therefore, there is a need for systems that improve the efficiency of regulatory compliance and risk management, thereby improving productivity and safety.

[0183] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0184] In this invention, the server includes a generation AI model means for providing appropriate countermeasures related to received laws and regulations and risks, a terminal means for inputting business content, a server means for receiving the input business content, and a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, thereby enabling prompt and appropriate responses to laws and regulations and risks.

[0185] The "terminal means for inputting work content" refers to a device that allows workers and managers in a factory to input work content and work procedures in voice or text format.

[0186] The "server means for receiving input business content" is a central processing unit for receiving data from the terminal into which the business content has been input, and for managing and processing the data.

[0187] "Natural language processing means for identifying relevant laws and potential risks from received business content" refers to a system that includes natural language processing (NLP) technology for analyzing the text data of received business content and identifying relevant laws and risks.

[0188] "Generative AI model means for generating countermeasures for identified risks" refers to an artificial intelligence (AI) model for generating appropriate countermeasures and guidelines for identified laws and regulations and risks.

[0189] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[0190] "Automatic sending means for automatically sending generated documents and emails to relevant parties" refers to a device or system for automatically sending generated documents and emails to appropriate relevant parties (e.g., factory managers or legal departments).

[0191] "Generative AI model means for providing appropriate countermeasures related to received laws and regulations and risks" is a generative AI model for providing preventive measures and countermeasures based on received laws and regulations and risks.

[0192] The "template selection means" is a system that has the function of selecting an appropriate template according to the identified risks and countermeasures.

[0193] A "feedback input means" is a device or system that allows a user to input feedback on the content of automatically generated documents or emails, and that allows that feedback to be reflected in improving the generative AI model.

[0194] The present invention provides a system for automating a series of processes for compliance with regulations and risk management, from identifying the work content of work robots in a factory to presenting appropriate countermeasures, automatically generating and sending documents, and collecting feedback. A detailed embodiment of this system will be described.

[0195] System configuration

[0196] The system consists of the following hardware and software means:

[0197] 1. Terminal means: A device used to input work content. Input can be done by voice or text. A specific example is the microphone and keyboard installed on a work robot in a factory.

[0198] 2. Server means: A central processing unit that receives and processes input business content. It has a built-in database and analytical model, and performs high-speed and accurate processing.

[0199] 3. Natural Language Processing (NLP): Analyzes business content and identifies relevant laws and regulations and potential risks. For example, a natural language processing library such as Spacy is used.

[0200] 4. Generative AI model means: Generate appropriate countermeasures to address identified risks. For example, the GPT-3 model using the OpenAI API.

[0201] 5. Automatic document generation means: This has the function of automatically creating the necessary documents and emails based on the generated countermeasures. A template-based document generation system is used.

[0202] 6. Automated Delivery: Automatically send generated documents and emails to relevant parties. This includes email systems and internal notification systems.

[0203] 7. Feedback input means: A system in which users can input feedback on the contents of automatically generated documents and emails, and that feedback is reflected in improving the generative AI model.

[0204] Program processing explanation

[0205] The server receives the business content from the terminal and analyzes the input content using natural language processing means. From the analyzed content, relevant laws and regulations and potential risks are identified, and based on that, a generative AI model provides countermeasures. Next, an automatic document generation means automatically generates documents and emails based on the countermeasures, which are then sent to relevant parties using an automatic sending means. The entire process aims to improve business efficiency and ensure swift and appropriate compliance with laws and regulations and risk management.

[0206] Examples of hardware and software used include:

[0207] Hardware: Factory robots, servers, smart devices for administrators

[0208] Software: Spacy (natural language processing library), OpenAI GPT-3 API (generative AI model), email system, template-based document generation system

[0209] Adding specific examples

[0210] For example, when introducing a new product assembly line in a factory, a manager inputs "Introducing a new product assembly line" into the robot's voice input system. The server receives this information and uses natural language processing to identify relevant laws and regulations, such as the Industrial Safety and Health Act, and potential risks, such as "Worker Safety Risks" and "Environmental Pollution Risks." The generative AI model then generates countermeasures based on the following prompt:

[0211] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[0212] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[0213] Please tell me the appropriate countermeasures.

[0214] Based on the proposed measures, the server automatically generates documents such as checklists and guidelines and sends them to the factory manager and legal department. The manager reviews the received documents and provides feedback as necessary, which the system uses to improve the system in the future.

[0215] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[0216] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0217] Step 1:

[0218] The user inputs the job details. A worker or manager in the factory inputs the specific job details (for example, "Introduce an assembly line for a new product") into the work robot's terminal using voice or text format. The input data is saved on the terminal and sent to the server.

[0219] Input: Task details (e.g., "Install an assembly line for a new product")

[0220] Output: Text data transmitted to the terminal, data sent to the server

[0221] Step 2:

[0222] The server receives the work content. The entered work content data is received and prepared for analysis. The received work content is saved to proceed to the next analysis stage.

[0223] Input: Text data sent from the terminal

[0224] Output: Data ready for analysis

[0225] Step 3:

[0226] The business content is analyzed using natural language processing. The server analyzes the received business content using a natural language processing library such as Spacy, and identifies relevant laws and potential risks. For example, relevant laws such as the "Occupational Safety and Health Act" and "Environmental Protection Act" and risks such as "worker safety risks" and "environmental pollution risks" are identified.

[0227] Input: Data ready for analysis

[0228] Output: Identified regulations and potential risks

[0229] Step 4:

[0230] Countermeasures are generated using a generative AI model. The server uses a generative AI model such as OpenAI GPT-3 to generate appropriate countermeasures for the identified laws and potential risks. For this, the following prompt sentences are used:

[0231] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[0232] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[0233] Please tell me the appropriate countermeasures.

[0234] Input: Identified laws and regulations and potential risks, prompt text

[0235] Output: Generated countermeasures

[0236] Step 5:

[0237] Documents and emails are automatically generated using an automatic document generation means. Based on the generated countermeasures, the server uses a template-based automatic document generation system to create the necessary documents and emails. For example, "checklists" and "guidelines" are created.

[0238] Input: Generated countermeasures

[0239] Output: Generated documents and emails

[0240] Step 6:

[0241] Send documents and emails to relevant parties using automated methods. The server automatically sends generated documents and emails to relevant parties, such as factory management or the legal department. This process uses email systems and internal notification systems.

[0242] Input: Generated documents and emails

[0243] Output: Documents and emails sent to stakeholders

[0244] Step 7:

[0245] The user enters feedback. Administrators and other relevant parties review the contents of received documents and emails and enter feedback as necessary. This feedback is sent to the server and used to improve the accuracy of the generative AI model.

[0246] Input: User feedback

[0247] Output: Feedback data stored on the server

[0248] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[0249] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0250] The system of the present invention is designed to enable even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and by adding a function to recognize and respond to the user's emotions, the content of documents and emails can be optimized. The following describes in detail the embodiments of the present invention.

[0251] System Overview

[0252] The system mainly consists of the following elements:

[0253] Terminal means for inputting business details

[0254] Server means for receiving input business details

[0255] Natural language processing means to identify relevant laws and potential risks from received business content

[0256] A generative AI model that generates countermeasures to address identified risks

[0257] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0258] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0259] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[0260] An emotion engine that recognizes emotions based on user input and feedback

[0261] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[0262] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[0263] Example of a system

[0264] Terminal means

[0265] The user inputs the details of the task in text format using the input form on the terminal. For example, the user may input "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0266] Server Means

[0267] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0268] Natural language processing and generative AI modeling tools

[0269] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0270] Automatic document generation method

[0271] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0272] Automatic transmission method

[0273] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0274] Feedback Input Method

[0275] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0276] Emotion engine and wording optimization

[0277] The emotion engine analyzes user input and feedback and recognizes the emotion. For example, if a user inputs an urgent task, the emotion engine recognizes this as the emotion representing "urgent." The recognized emotion is used by the wording optimization tool to optimize the tone and wording of generated documents and emails. This allows recipients to better understand the content and respond quickly.

[0278] Emotion data reflection method

[0279] The emotion data recognized by the emotion engine is also reflected in the learning data of the generative AI model, improving the model's accuracy and flexibility, and enabling more appropriate responses in future business content analysis and document generation.

[0280] Specific examples

[0281] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[0282] 1. The user enters the details of the job into the terminal and sends it to the server.

[0283] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[0284] 3. The server requests the generative AI model to generate risk countermeasures and receives the appropriate countermeasures.

[0285] 4. The server automatically generates documents and emails based on the corrective action.

[0286] 5. Language optimization tools adjust the tone of documents and emails based on user sentiment.

[0287] 6. The server automatically sends the generated document to the marketing and legal departments.

[0288] 7. The user checks the results on the device and provides feedback if necessary.

[0289] 8. The feedback content is sent to the server, where it is emotionally analyzed by the emotion engine and reflected in the training data of the generative AI model.

[0290] In this way, the present invention is highly effective in improving business efficiency and risk management by automating the generation and transmission of legal documents that reflect the user's feelings.

[0291] The processing flow will be explained below.

[0292] Step 1:

[0293] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[0294] Step 2:

[0295] The device sends the entered business details to the server using an API, which transfers the data securely.

[0296] Step 3:

[0297] The server receives the submitted business details and passes the received data to the natural language processing engine.

[0298] Step 4:

[0299] The server uses a natural language processing engine to analyze the received business content. As a result of the analysis, the structure and meaning of the sentence are understood, and relevant laws and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" are identified.

[0300] Step 5:

[0301] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[0302] Step 6:

[0303] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[0304] Step 7:

[0305] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[0306] Step 8:

[0307] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[0308] Step 9:

[0309] The user enters feedback from the terminal, providing additional information and suggestions for improvement based on the content of the generated documents and emails.

[0310] Step 10:

[0311] The server receives the feedback and passes it to the emotion engine for analysis. The user's emotion is identified from the feedback content. For example, the emotion meaning "urgency" is recognized.

[0312] Step 11:

[0313] The server optimizes the wording of documents and emails based on the emotional data from the emotion engine, adjusting the tone and wording as needed.

[0314] Step 12:

[0315] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[0316] Step 13:

[0317] The server reflects the emotion data identified by the emotion engine in the learning data of the generative AI model, improving the accuracy of the generative AI model and making future analysis and generation more accurate.

[0318] In this way, the system automates a series of processes, from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, improving accuracy through feedback, and even optimizing wording by combining an emotion engine. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, improving work efficiency and risk management.

[0319] Example 2

[0320] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0321] There is a demand for a system that can respond to legal issues quickly and appropriately, even when employees are not familiar with legal matters. It is also necessary to optimize the content of documents and emails by reflecting the user's feelings, and ensure smooth correspondence between the parties involved. Conventional systems do not automate the identification of laws and regulations or risk countermeasures, which means that legal responses take time, and the system does not reflect the user's feelings, making it difficult to communicate appropriately.

[0322] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0323] In this invention, the server includes a terminal means for inputting business content, a server means for receiving the input business content, a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, a generative AI model means for generating countermeasures to address the identified risks, an automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, an automatic sending means for automatically sending the generated documents and emails to relevant parties, a feedback input means for allowing a user to input feedback on the content of the automatically generated documents and emails and reflecting this feedback in improving the accuracy of the generative AI model, an emotion engine for recognizing a user's emotions based on the user's input content and feedback, a wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine, and an emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means. This enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and documents and emails that reflect the user's emotions can be generated and sent, thereby facilitating communication.

[0324] The "terminal means for inputting business content" is a device equipped with an input device and an input form for a user to input business content in text format.

[0325] The "server means for receiving the input business content" is a server device for receiving the business content input by the user through the terminal means.

[0326] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to means that uses natural language processing technology to analyze input business content and identify relevant laws and regulations and potential risks.

[0327] A "generative AI model means for generating countermeasures for identified risks" is a means for using an artificial intelligence model to generate appropriate countermeasures for identified risks.

[0328] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a means for automatically creating necessary documents and e-mails based on the generated solutions.

[0329] The "automatic sending means for automatically sending the generated document or e-mail to the relevant person" is a means for automatically sending the generated document or e-mail to the designated relevant person.

[0330] "Feedback input means that allows users to input feedback on the content of automatically generated documents or emails and reflect it in improving the accuracy of the generative AI model" refers to a means that allows users to input opinions and suggestions for improvement on the content of automatically generated documents or emails, and uses that feedback information as learning data for the generative AI model.

[0331] "Emotion engine that recognizes user emotions based on user input and feedback" is an engine that analyzes and recognizes emotions from the content and feedback entered by the user.

[0332] The "wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine" is a means for optimally adjusting the content of generated documents and emails based on the user's emotions recognized by the emotion engine.

[0333] The "emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means" is a means for incorporating the emotion data recognized by the emotion engine into the learning data of the generative AI model.

[0334] The "template selection means" is a means for selecting an appropriate template for a document or email based on the identified risks and countermeasures.

[0335] The system of the present invention is designed to process legal work quickly and appropriately, and enables the generation of documents and emails that reflect the user's feelings. A specific embodiment of the present invention will be described below.

[0336] 1. System Configuration

[0337] The system consists of the following elements:

[0338] Terminal means for inputting business details

[0339] Server means for receiving input business details

[0340] Natural language processing means to identify relevant laws and potential risks from received business content

[0341] A generative AI model that generates countermeasures to address identified risks

[0342] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0343] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0344] A feedback input method that allows users to input feedback on the content of automatically generated documents and emails, and reflects this feedback in improving the accuracy of the generative AI model.

[0345] An emotion engine that recognizes user emotions based on user input and feedback

[0346] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[0347] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[0348] 2. Hardware and Software Used

[0349] Terminal means: The user uses an input device such as a computer or tablet to enter the details of the work in text format into a dedicated input form.

[0350] Server means: The server is a platform for receiving and processing business content sent from the terminal. The server may also use cloud-based services.

[0351] Natural language processing: For natural language processing, we use NLP libraries such as SpaCy and BERT to analyze the input business content and identify relevant laws and regulations and potential risks.

[0352] Generative AI model means: To generate countermeasures, we use generative AI models such as GPT-3 and BERT, which automatically generate appropriate countermeasures.

[0353] Automatic document generation method: To automatically generate documents and emails, a specified template is used. The template is automatically selected according to the business content and the solution.

[0354] Emotion Engine: For emotion recognition, emotion analysis algorithms (e.g., TextBlob, VADER) are used to identify emotions from user input and feedback.

[0355] 3. Specific Examples

[0356] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[0357] 1. Using the terminal means, the user enters "Plan a marketing strategy for new product A" into the input form and presses the send button.

[0358] 2. The terminal sends the entered business details to the server.

[0359] 3. The server analyzes the received data using natural language processing tools (e.g., SpaCy) to identify consumer protection laws and advertising regulations.

[0360] 4. The server sends a prompt to the generative AI model saying, "Please generate solutions to address the risk of violating advertising regulations," and receives the solutions.

[0361] 5. The server automatically generates a marketing plan using a template based on the solution.

[0362] 6. The server automatically sends the generated plan to the marketing and legal departments.

[0363] 7. The user checks the plan and enters feedback such as "Please correct this part."

[0364] 8. The server analyzes the feedback, recognizes the emotion of "urgency," and adjusts the tone of the document using language optimization techniques.

[0365] 9. The server reflects the recognized emotion data in the learning data of the generative AI model, aiming to improve accuracy from the next time onwards.

[0366] Example prompt sentence:

[0367] "Identify the laws and regulations and potential risks that need to be considered when developing a marketing strategy for new product A, and generate appropriate countermeasures."

[0368] "Generate appropriate documentation to comply with consumer protection laws and advertising regulations when developing your marketing strategies."

[0369] In this way, the present invention enables even users who are not familiar with legal matters to respond to legal matters quickly and appropriately, and facilitates smooth communication through the creation and transmission of documents and emails that reflect emotions.

[0370] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0371] Step 1:

[0372] The user uses the terminal means to input the details of the job in text format into the input form. For example, the user might input "Plan a marketing strategy for new product A" and press the send button.

[0373] Input: Job description (e.g., "Plan a marketing strategy for new product A")

[0374] Output: Text data of the work content is sent from the terminal

[0375] Step 2:

[0376] The terminal transmits the text data of the entered business details to the server.

[0377] Input: Text data of business content

[0378] Output: Data sent to the server

[0379] Step 3:

[0380] The server analyzes the received text data of the business content using natural language processing tools (e.g., SpaCy). As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" may be identified.

[0381] Input: Text data of business content

[0382] Data processing: Applying natural language processing to identify relevant laws and risks

[0383] Output: List of relevant laws and regulations and potential risks

[0384] Step 4:

[0385] The server requests the generative AI model to generate a solution based on the identified laws and risks. For example, it sends a prompt saying, "Please generate a solution to address the risk of violating advertising display regulations." The generative AI model generates an appropriate solution and sends it back to the server.

[0386] Input: List of relevant laws and regulations and potential risks, prompt text

[0387] Data computation: Using generative AI models to generate solutions

[0388] Output: A list of appropriate actions

[0389] Step 5:

[0390] Based on the received solutions, the server uses an automatic document generation means to automatically generate the necessary documents and emails. At this time, a template appropriate for the business content and solutions is selected, and specific documents and emails are created. For example, a "marketing plan" or "compliance checklist" is automatically generated.

[0391] Input: Solution list, template

[0392] Data processing: Generate documents and emails based on solutions and templates

[0393] Output: Generated documents and emails

[0394] Step 6:

[0395] The server automatically sends the generated documents and emails to the appropriate parties, including the user, relevant internal departments (e.g., marketing, legal), and external legal counsel.

[0396] Input: Generated documents and emails

[0397] Data processing: Send using email transmission protocols (e.g., SMTP)

[0398] Output: Documents and emails sent to stakeholders

[0399] Step 7:

[0400] The user checks the content of the automatically generated document or email and enters feedback as needed. For example, the user may enter feedback such as "Please correct this part." The feedback is sent from the terminal to the server.

[0401] Input: Feedback

[0402] Output: Feedback is sent to the server

[0403] Step 8:

[0404] The server receives the feedback and uses an emotion engine to analyze the user's emotions. For example, the emotion representing "urgency" is recognized. Based on this emotion, a wording optimizer adjusts the tone and wording of the generated document or email.

[0405] Input: Feedback

[0406] Data Computation: Emotion Analysis with Emotion Engine

[0407] Output: Recognized sentiment, optimized wording

[0408] Step 9:

[0409] The server reflects the emotion data recognized by the emotion engine in the learning data of the generative AI model, enabling more appropriate responses in subsequent analyses and document generation.

[0410] Input: Emotion data

[0411] Data processing: Reflecting emotion data in the training data of the generative AI model

[0412] Output: Updated training data for the generative AI model

[0413] In this way, a system is realized in which the processing at each step is linked, allowing even users unfamiliar with legal matters to respond quickly and appropriately to legal matters.

[0414] (Application example 2)

[0415] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0416] With conventional systems, it was difficult for employees without legal expertise to generate documents that complied with regulations, which increased the likelihood of mistakes and risks. Furthermore, the system was unable to respond flexibly to user sentiment, resulting in reduced operational efficiency. These issues must be resolved, particularly for online shopping sites, where compliance with regulations and communication based on user sentiment are crucial.

[0417] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes server means for receiving input business content, natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, generation AI model means for generating countermeasures for addressing the identified risks, automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, automatic sending means for automatically sending the generated documents and emails to relevant parties, emotion analysis means for recognizing emotions from user input content and feedback, and wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means. This enables appropriate responses to laws and regulations and flexible document generation based on user emotions.

[0418] The "terminal means for inputting business details" is a device equipped with an interface for a user to input business details in text format.

[0419] The "server means for receiving input business content" is a server for receiving and storing data sent by a user from a terminal means.

[0420] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to natural language processing technology for analyzing received text data and identifying relevant laws and regulations and potential risks.

[0421] A "generative AI model means for generating countermeasures to address identified risks" is an artificial intelligence model that generates methods and guidelines for addressing identified risks.

[0422] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[0423] "Automatic sending means for automatically sending generated documents and e-mails to relevant parties" refers to a system that has the function of automatically sending generated documents and e-mails to designated relevant parties.

[0424] "Emotion analysis means for recognizing emotions from user input and feedback" is a technology for determining emotions from text data and feedback entered by the user.

[0425] The "wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means" is a system that adjusts the tone and expression of generated documents and emails based on the emotions recognized by the emotion analysis means.

[0426] The "template selection means for selecting a document or email template based on identified risks and countermeasures" is a system that has the function of automatically selecting the most appropriate template based on risks and countermeasures.

[0427] "Feedback input means that allows users to input feedback on the contents of automatically generated documents and emails, and reflect this in improving the accuracy of the generative AI model" refers to a system that has the function of allowing users to provide feedback on automatically generated documents and emails, and to improve the generative AI model based on that feedback.

[0428] "A means for updating a generative AI model that reflects feedback content and emotional data based on generated documents and emails in the learning data of the generative AI model" is a technology that uses provided feedback and emotional data to update a generative AI model and improve the accuracy and flexibility of the model.

[0429] This embodiment of the present invention relates to a compliance assistance system for online shopping sites. The purpose of this system is to quickly and accurately perform legal checks when users release new products or services. A specific embodiment of this system will be described below.

[0430] System configuration

[0431] This system mainly consists of the following hardware and software:

[0432] Hardware: Smartphone

[0433] Software: Python, natural language processing libraries (spaCy, NLTK), sentiment analysis library (TextBlob), generative AI model (OpenAI GPT-3), Flask, REST API, Jinja2, smtplib

[0434] Program processing

[0435] 1. Input form: Using a smartphone app, users enter text descriptions of new products and services. This input form is built using HTML and JavaScript.

[0436] 2. Data reception: The data entered by the user is sent to the server via a REST API built using Flask. The server receives and stores this data.

[0437] 3. Natural Language Analysis: The server analyzes the received input data using Python and natural language processing libraries (spaCy, NLTK). This analysis identifies relevant laws and regulations and potential risks.

[0438] 4. Generative AI model: Based on the identified risks, a generative AI model (GPT-3) is used to generate countermeasures. The generative AI model is invoked using Python and the OpenAI API.

[0439] 5. Sentiment Analysis: TextBlob is used to analyze sentiment from user text input and feedback. This sentiment data is used to generate documents and emails.

[0440] 6. Automatic document generation: Based on the generated responses and the results of sentiment analysis, the necessary documents and emails are automatically generated using Jinja2.

[0441] 7. Automatic sending: Generated documents and emails are automatically sent to the relevant parties using Python's smtplib library.

[0442] 8. Feedback collection: Users can input feedback on the content of generated documents and emails through a smartphone app. This feedback is sent back to the server and reflected in the training data for the generative AI model.

[0443] In this way, appropriate measures for legal regulations and flexible document generation based on user feelings are realized.

[0444] Specific examples

[0445] For example, the user enters the following prompt text:

[0446] Develop a marketing strategy for new product A, including how to effectively appeal to the target market while complying with consumer protection laws and advertising regulations.

[0447] The server that receives this input uses natural language processing means to identify relevant laws and regulations, such as "consumer protection laws" and "regulations on advertising display," as well as risks. Then, the generative AI model means generates countermeasures for these risks. Based on the generated countermeasures, the automatic document generation means creates appropriate marketing plans and compliance checklists.

[0448] The sentiment analysis means analyzes the emotion of "effectively appealing" that the user emphasized in their input, and the wording optimization means optimizes the tone and expression of the document. Finally, the generated document or email is automatically sent to the relevant department or person in charge. The user can review the generated document and provide feedback as needed, which will improve the accuracy and responsiveness of the generation AI model from the next time onwards. In this way, business efficiency and risk management are improved.

[0449] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0450] Step 1:

[0451] User enters business details

[0452] A user enters a text description of a new product or service into an input form on a smartphone app. For example, they might enter, "I'm planning a marketing strategy for new product A. I want to include ways to effectively appeal to the target market while also complying with consumer protection laws and advertising regulations." This input data is sent from an input form built using HTML and JavaScript.

[0453] Step 2:

[0454] Data Receipt and Storage

[0455] Data on work details sent from the smartphone app is sent to the server via a REST API using Flask. The server receives this data and stores it in a database. An example of input data is text such as "Plan a marketing strategy for new product A."

[0456] Step 3:

[0457] natural language analysis

[0458] The server analyzes the received business text using natural language processing libraries (spaCy, NLTK). This process identifies relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display" are identified. The input data is text, and the output data is the name of the law and risk information.

[0459] Step 4:

[0460] Generative AI model generates solutions

[0461] The server uses a generative AI model (GPT-3) to generate appropriate countermeasures based on the identified risks. The generative AI model operates through Python and the OpenAI API, receiving legal names and risk information as input data. The output data is specific countermeasures and guidelines.

[0462] Step 5:

[0463] Emotion analysis

[0464] The server uses TextBlob to analyze the sentiment from the user's input text and feedback. This sentiment analysis extracts the user's emphasis and sentiment. The input data is text, and the output data is sentiment information such as positive, negative, and urgency.

[0465] Step 6:

[0466] Auto-generated documents

[0467] The server uses Jinja2 to automatically generate the necessary documents and emails based on the generated responses and the results of sentiment analysis. The input data are responses and sentiment data, and the output data is customized documents and emails based on templates. For example, a marketing plan or legal checklist based on regulations is generated.

[0468] Step 7:

[0469] Automatic transmission

[0470] The created documents and emails are automatically sent to the relevant parties using the Python smtplib library. The input data is the created document or email, and the output data is recorded on the server as a notification of the completion of sending.

[0471] Step 8:

[0472] Feedback collection

[0473] Users input feedback on generated documents and emails through a smartphone app. The feedback is sent to the server and stored in the database again. This data is used as training data for the generative AI model, helping to improve the accuracy of future document generation. The input data is the feedback content, and the output data is the improved response of the generative AI model.

[0474] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0475] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0476] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0477] [Second embodiment]

[0478] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0479] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0480] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0481] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0482] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0483] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0484] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0485] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0486] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0487] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0488] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0489] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0490] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[0491] System Overview

[0492] The system mainly consists of the following elements:

[0493] Terminal means for inputting business details

[0494] Server means for receiving input business details

[0495] Natural language processing means to identify relevant laws and potential risks from received business content

[0496] A generative AI model that generates countermeasures to address identified risks

[0497] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0498] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0499] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[0500] Example of a system

[0501] Terminal means

[0502] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0503] Server Means

[0504] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0505] Natural language processing and generative AI modeling tools

[0506] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0507] Automatic document generation method

[0508] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0509] Automatic transmission method

[0510] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0511] Feedback Input Method

[0512] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0513] Specific examples

[0514] For example, if a user enters "Plan a marketing strategy for new product A," the system will automatically go through the following steps to identify relevant laws and regulations, assess risks, present countermeasures, and generate and send documents.

[0515] 1. The user enters the details of the job into the terminal and sends it to the server.

[0516] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[0517] 3. The server uses a generative AI model to generate countermeasures for the identified risks.

[0518] 4. The server automatically generates appropriate documents and emails based on the corrective action.

[0519] 5. The server automatically sends the generated document to the marketing and legal departments.

[0520] 6. The user reviews the document on their device and provides feedback if necessary.

[0521] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[0522] The processing flow will be explained below.

[0523] Step 1:

[0524] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[0525] Step 2:

[0526] The device sends the entered business details to the server, and the data is securely transferred using an API.

[0527] Step 3:

[0528] The server receives the submitted business content and passes the received data to a natural language processing (NLP) engine for analysis.

[0529] Step 4:

[0530] The server uses a natural language processing engine to analyze the received business content, understand the structure and meaning of the sentences, and identify relevant laws and regulations and potential risks.

[0531] Step 5:

[0532] The server identifies relevant laws and regulations and potential risks. For example, it identifies "consumer protection laws" and "regulations on advertising display" and recognizes the risk of violations.

[0533] Step 6:

[0534] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[0535] Step 7:

[0536] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[0537] Step 8:

[0538] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[0539] Step 9:

[0540] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[0541] Step 10:

[0542] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[0543] Step 11:

[0544] The user checks the contents of documents generated on the device and emails sent, and confirms that the contents are appropriate.

[0545] Step 12:

[0546] Users can input feedback as needed, which is sent from the device to the server and used to improve the accuracy of the generative AI model.

[0547] In this way, the system automates everything from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, and improving accuracy through feedback. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, resulting in improved work efficiency and risk management.

[0548] Example 1

[0549] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0550] In the past, it was difficult for employees with insufficient legal knowledge to respond quickly and appropriately to legal issues. Furthermore, the entire process of identifying relevant laws and potential risks, developing countermeasures, and creating and sending documents required a great deal of time and effort. Furthermore, feedback on the content of generated documents and emails was not properly reflected, making it difficult to improve the accuracy of the system. This resulted in reduced work efficiency and inadequate risk management.

[0551] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0552] In this invention, the server includes a means for receiving business content, a means for identifying relevant laws and regulations and potential risks from the received business content, and a generative AI model means for generating countermeasures to address the identified risks. This allows even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters. Furthermore, by including a means for automatically generating necessary documents and emails based on the generated countermeasures, a means for automatically sending the generated documents and emails to relevant parties, and a means for users to input feedback on the content of the automatically generated documents and emails and reflect this feedback in improving the accuracy of the generative AI model, it is possible to further improve business efficiency and the accuracy of risk management.

[0553] "Business content" refers to text information related to a business that a user inputs into a terminal.

[0554] "Terminal means" refers to a device or its interface that allows a user to input business details.

[0555] The "server means" is a server that receives the business contents sent from the terminal and performs the processing.

[0556] "Natural language processing means" refers to technologies and algorithms used to analyze received business content and identify relevant laws and regulations and potential risks.

[0557] "Generative AI model means" refers to an artificial intelligence model for generating countermeasures to identified risks.

[0558] The "automatic document generation means" is a system or program for automatically creating the necessary documents or emails based on the generated solutions.

[0559] "Automatic sending means" refers to a system or method for automatically sending generated documents or emails to relevant parties.

[0560] "Feedback input means" refers to a system or method that allows users to input feedback on the contents of automatically generated documents or emails, and uses that feedback to improve the accuracy of the generative AI model.

[0561] "Template selector" means a system or algorithm for selecting document or email templates based on identified risks and countermeasures.

[0562] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[0563] System Configuration

[0564] The system includes the following major hardware and software elements:

[0565] Terminal means: A device for inputting work details. For example, a PC or tablet. The user inputs the work details and sends them to the server.

[0566] Server means: A server that receives and processes the business content sent from the terminal. Specifically, it runs natural language processing (NLP) libraries (e.g., SpaCy, BERT) and generative AI models (e.g., GPT-3).

[0567] Natural language processing means: A program that runs on the server and analyzes the received business content to identify relevant laws and regulations and potential risks.

[0568] Generative AI model means: An artificial intelligence model integrated into the server generates countermeasures for identified risks.

[0569] Automatic document generation: Automatically create documents and emails based on the solutions generated on the server. A template engine (e.g., Jinja2) is used.

[0570] Automatic sending means: A system that sends generated documents and emails to the relevant parties, for example, via an SMTP server.

[0571] Feedback input method: The user checks the contents of the generated documents and emails, and inputs feedback as necessary to improve the accuracy of the generative AI model. The feedback data is stored in a database (e.g., PostgreSQL, MySQL).

[0572] Template selection tools: Tools for selecting document and email templates based on identified risks and treatments.

[0573] Example of a system

[0574] Terminal means

[0575] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0576] Server Means

[0577] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0578] Natural language processing and generative AI modeling tools

[0579] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0580] Automatic document generation method

[0581] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0582] Automatic transmission method

[0583] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0584] Feedback Input Method

[0585] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0586] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[0587] Specific examples

[0588] For example, if a user inputs "Plan a marketing strategy for new product A," the process will go through the following steps:

[0589] 1. The business details entered by the user are sent to the server from the terminal.

[0590] 2. The server receives the business details and performs natural language processing to identify relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display."

[0591] 3. The server uses the generative AI model to generate a solution, such as suggesting "create advertising copy that complies with the Consumer Protection Act."

[0592] 4. The server automatically generates documents and emails based on the measures taken, such as a "marketing plan" or a "compliance checklist."

[0593] 5. The server automatically sends the generated document to the relevant parties.

[0594] 6. The user reviews the document and provides feedback if necessary.

[0595] Examples of prompts

[0596] An example of a prompt to input to a generative AI model is: "Analyze the laws and potential risks related to the marketing strategy for new product A, and propose appropriate countermeasures."

[0597] Based on this prompt, the generative AI model suggests appropriate actions, and the system then generates the necessary documents and emails.

[0598] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0599] Step 1:

[0600] The user inputs the details of the job into the terminal and sends it to the server.

[0601] Input: Enter the job description in text format as "Plan a marketing strategy for new product A."

[0602] Output: Text data of the work content sent from the terminal to the server.

[0603] Specific operation: The user enters "Plan a marketing strategy for new product A" into the input form on the terminal and clicks the submit button.

[0604] Step 2:

[0605] The server receives the business content and analyzes it using natural language processing means.

[0606] Input: Text data of the work content sent from the terminal.

[0607] Output: Data on relevant laws and regulations and potential risks extracted from business operations.

[0608] Specific operation: The server analyzes the text received from the device, "Plan a marketing strategy for new product A," using a natural language processing tool (such as SpaCy or BERT), and identifies relevant laws and regulations such as "consumer protection laws" and "regulations on advertising display," as well as potential risks.

[0609] Step 3:

[0610] The server uses the generative AI model to generate countermeasures for the identified risks.

[0611] Input: Data on relevant laws and regulations and potential risks identified through natural language processing.

[0612] Output: Text data of relevant laws and regulations and countermeasures for potential risks.

[0613] Specific operation: When a risk related to the Consumer Protection Act is identified, the server sends a prompt to the generative AI model (e.g., GPT-3) saying, "Please create advertising text that complies with the Consumer Protection Act," and generates a solution (appropriate advertising text and display method).

[0614] Step 4:

[0615] The server automatically generates the necessary documents and emails based on the generated solutions.

[0616] Input: Text data of solutions obtained from the generative AI model.

[0617] Output: Auto-generated documents and email data.

[0618] Specific operation: The server receives the generated solutions in text format and uses a template engine (e.g., Jinja2) to automatically generate a "marketing plan in accordance with consumer protection laws" and a "compliance checklist."

[0619] Step 5:

[0620] The server automatically sends generated documents and emails to the relevant parties.

[0621] Input: Data from automatically generated documents and emails.

[0622] Output: The sent document or email reaches the relevant person.

[0623] Specific operation: The server sends the generated "marketing plan" via email to the user, marketing department, and legal department via the SMTP server.

[0624] Step 6:

[0625] The user checks the generated documents and emails and provides feedback.

[0626] Input: The contents of the document or email sent.

[0627] Output: Feedback data from users.

[0628] Specific actions: The user reviews the received "Marketing Plan" and submits feedback by entering a comment such as "This ad copy is appropriate, but I would like it to be a little more specific."

[0629] Step 7:

[0630] The server receives user feedback and uses it to improve the accuracy of the generative AI model.

[0631] Input: Feedback data submitted by the user.

[0632] Output: Updated generative AI model data.

[0633] Specific operation: The server stores the received feedback in a database and uses the feedback data as training data for the generative AI model to improve the accuracy of the model.

[0634] (Application example 1)

[0635] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0636] Conventional regulatory compliance and risk management systems have had difficulty identifying work content and providing appropriate countermeasures quickly and accurately. In particular, with regard to work robots in factories, immediate responses to complex regulations and risks are required, but if on-site personnel lack sufficient legal knowledge, responses may be delayed. Therefore, there is a need for systems that improve the efficiency of regulatory compliance and risk management, thereby improving productivity and safety.

[0637] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0638] In this invention, the server includes a generation AI model means for providing appropriate countermeasures related to received laws and regulations and risks, a terminal means for inputting business content, a server means for receiving the input business content, and a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, thereby enabling prompt and appropriate responses to laws and regulations and risks.

[0639] The "terminal means for inputting work content" refers to a device that allows workers and managers in a factory to input work content and work procedures in voice or text format.

[0640] The "server means for receiving input business content" is a central processing unit for receiving data from the terminal into which the business content has been input, and for managing and processing the data.

[0641] "Natural language processing means for identifying relevant laws and potential risks from received business content" refers to a system that includes natural language processing (NLP) technology for analyzing the text data of received business content and identifying relevant laws and risks.

[0642] "Generative AI model means for generating countermeasures for identified risks" refers to an artificial intelligence (AI) model for generating appropriate countermeasures and guidelines for identified laws and regulations and risks.

[0643] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[0644] "Automatic sending means for automatically sending generated documents and emails to relevant parties" refers to a device or system for automatically sending generated documents and emails to appropriate relevant parties (e.g., factory managers or legal departments).

[0645] "Generative AI model means for providing appropriate countermeasures related to received laws and regulations and risks" is a generative AI model for providing preventive measures and countermeasures based on received laws and regulations and risks.

[0646] The "template selection means" is a system that has the function of selecting an appropriate template according to the identified risks and countermeasures.

[0647] A "feedback input means" is a device or system that allows a user to input feedback on the content of automatically generated documents or emails, and that allows that feedback to be reflected in improving the generative AI model.

[0648] The present invention provides a system for automating a series of processes for compliance with regulations and risk management, from identifying the work content of work robots in a factory to presenting appropriate countermeasures, automatically generating and sending documents, and collecting feedback. A detailed embodiment of this system will be described.

[0649] System configuration

[0650] The system consists of the following hardware and software means:

[0651] 1. Terminal means: A device used to input work content. Input can be done by voice or text. A specific example is the microphone and keyboard installed on a work robot in a factory.

[0652] 2. Server means: A central processing unit that receives and processes input business content. It has a built-in database and analytical model, and performs high-speed and accurate processing.

[0653] 3. Natural Language Processing (NLP): Analyzes business content and identifies relevant laws and regulations and potential risks. For example, a natural language processing library such as Spacy is used.

[0654] 4. Generative AI model means: Generate appropriate countermeasures to address identified risks. For example, the GPT-3 model using the OpenAI API.

[0655] 5. Automatic document generation means: This has the function of automatically creating the necessary documents and emails based on the generated countermeasures. A template-based document generation system is used.

[0656] 6. Automated Delivery: Automatically send generated documents and emails to relevant parties. This includes email systems and internal notification systems.

[0657] 7. Feedback input means: A system in which users can input feedback on the contents of automatically generated documents and emails, and that feedback is reflected in improving the generative AI model.

[0658] Program processing explanation

[0659] The server receives the business content from the terminal and analyzes the input content using natural language processing means. From the analyzed content, relevant laws and regulations and potential risks are identified, and based on that, a generative AI model provides countermeasures. Next, an automatic document generation means automatically generates documents and emails based on the countermeasures, which are then sent to relevant parties using an automatic sending means. The entire process aims to improve business efficiency and ensure swift and appropriate compliance with laws and regulations and risk management.

[0660] Examples of hardware and software used include:

[0661] Hardware: Factory robots, servers, smart devices for administrators

[0662] Software: Spacy (natural language processing library), OpenAI GPT-3 API (generative AI model), email system, template-based document generation system

[0663] Adding specific examples

[0664] For example, when introducing a new product assembly line in a factory, a manager inputs "Introducing a new product assembly line" into the robot's voice input system. The server receives this information and uses natural language processing to identify relevant laws and regulations, such as the Industrial Safety and Health Act, and potential risks, such as "Worker Safety Risks" and "Environmental Pollution Risks." The generative AI model then generates countermeasures based on the following prompt:

[0665] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[0666] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[0667] Please tell me the appropriate countermeasures.

[0668] Based on the proposed measures, the server automatically generates documents such as checklists and guidelines and sends them to the factory manager and legal department. The manager reviews the received documents and provides feedback as necessary, which the system uses to improve the system in the future.

[0669] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[0670] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0671] Step 1:

[0672] The user inputs the job details. A worker or manager in the factory inputs the specific job details (for example, "Introduce an assembly line for a new product") into the work robot's terminal using voice or text format. The input data is saved on the terminal and sent to the server.

[0673] Input: Task details (e.g., "Install an assembly line for a new product")

[0674] Output: Text data transmitted to the terminal, data sent to the server

[0675] Step 2:

[0676] The server receives the work content. The entered work content data is received and prepared for analysis. The received work content is saved to proceed to the next analysis stage.

[0677] Input: Text data sent from the terminal

[0678] Output: Data ready for analysis

[0679] Step 3:

[0680] The business content is analyzed using natural language processing. The server analyzes the received business content using a natural language processing library such as Spacy, and identifies relevant laws and potential risks. For example, relevant laws such as the "Occupational Safety and Health Act" and "Environmental Protection Act" and risks such as "worker safety risks" and "environmental pollution risks" are identified.

[0681] Input: Data ready for analysis

[0682] Output: Identified regulations and potential risks

[0683] Step 4:

[0684] Countermeasures are generated using a generative AI model. The server uses a generative AI model such as OpenAI GPT-3 to generate appropriate countermeasures for the identified laws and potential risks. For this, the following prompt sentences are used:

[0685] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[0686] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[0687] Please tell me the appropriate countermeasures.

[0688] Input: Identified laws and regulations and potential risks, prompt text

[0689] Output: Generated countermeasures

[0690] Step 5:

[0691] Documents and emails are automatically generated using an automatic document generation means. Based on the generated countermeasures, the server uses a template-based automatic document generation system to create the necessary documents and emails. For example, "checklists" and "guidelines" are created.

[0692] Input: Generated countermeasures

[0693] Output: Generated documents and emails

[0694] Step 6:

[0695] Send documents and emails to relevant parties using automated methods. The server automatically sends generated documents and emails to relevant parties, such as factory management or the legal department. This process uses email systems and internal notification systems.

[0696] Input: Generated documents and emails

[0697] Output: Documents and emails sent to stakeholders

[0698] Step 7:

[0699] The user enters feedback. Administrators and other relevant parties review the contents of received documents and emails and enter feedback as necessary. This feedback is sent to the server and used to improve the accuracy of the generative AI model.

[0700] Input: User feedback

[0701] Output: Feedback data stored on the server

[0702] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[0703] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0704] The system of the present invention is designed to enable even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and by adding a function to recognize and respond to the user's emotions, the content of documents and emails can be optimized. The following describes in detail the embodiments of the present invention.

[0705] System Overview

[0706] The system mainly consists of the following elements:

[0707] Terminal means for inputting business details

[0708] Server means for receiving input business details

[0709] Natural language processing means to identify relevant laws and potential risks from received business content

[0710] A generative AI model that generates countermeasures to address identified risks

[0711] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0712] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0713] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[0714] An emotion engine that recognizes emotions based on user input and feedback

[0715] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[0716] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[0717] Example of a system

[0718] Terminal means

[0719] The user inputs the details of the task in text format using the input form on the terminal. For example, the user may input "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0720] Server Means

[0721] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0722] Natural language processing and generative AI modeling tools

[0723] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0724] Automatic document generation method

[0725] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0726] Automatic transmission method

[0727] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0728] Feedback Input Method

[0729] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0730] Emotion engine and wording optimization

[0731] The emotion engine analyzes user input and feedback and recognizes the emotion. For example, if a user inputs an urgent task, the emotion engine recognizes this as the emotion representing "urgent." The recognized emotion is used by the wording optimization tool to optimize the tone and wording of generated documents and emails. This allows recipients to better understand the content and respond quickly.

[0732] Emotion data reflection method

[0733] The emotion data recognized by the emotion engine is also reflected in the learning data of the generative AI model, improving the model's accuracy and flexibility, and enabling more appropriate responses in future business content analysis and document generation.

[0734] Specific examples

[0735] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[0736] 1. The user enters the details of the job into the terminal and sends it to the server.

[0737] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[0738] 3. The server requests the generative AI model to generate risk countermeasures and receives the appropriate countermeasures.

[0739] 4. The server automatically generates documents and emails based on the corrective action.

[0740] 5. Language optimization tools adjust the tone of documents and emails based on user sentiment.

[0741] 6. The server automatically sends the generated document to the marketing and legal departments.

[0742] 7. The user checks the results on the device and provides feedback if necessary.

[0743] 8. The feedback content is sent to the server, where it is emotionally analyzed by the emotion engine and reflected in the training data of the generative AI model.

[0744] In this way, the present invention is highly effective in improving business efficiency and risk management by automating the generation and transmission of legal documents that reflect the user's feelings.

[0745] The processing flow will be explained below.

[0746] Step 1:

[0747] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[0748] Step 2:

[0749] The device sends the entered business details to the server using an API, which transfers the data securely.

[0750] Step 3:

[0751] The server receives the submitted business details and passes the received data to the natural language processing engine.

[0752] Step 4:

[0753] The server uses a natural language processing engine to analyze the received business content. As a result of the analysis, the structure and meaning of the sentence are understood, and relevant laws and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" are identified.

[0754] Step 5:

[0755] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[0756] Step 6:

[0757] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[0758] Step 7:

[0759] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[0760] Step 8:

[0761] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[0762] Step 9:

[0763] The user enters feedback from the terminal, providing additional information and suggestions for improvement based on the content of the generated documents and emails.

[0764] Step 10:

[0765] The server receives the feedback and passes it to the emotion engine for analysis. The user's emotion is identified from the feedback content. For example, the emotion meaning "urgency" is recognized.

[0766] Step 11:

[0767] The server optimizes the wording of documents and emails based on the emotional data from the emotion engine, adjusting the tone and wording as needed.

[0768] Step 12:

[0769] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[0770] Step 13:

[0771] The server reflects the emotion data identified by the emotion engine in the learning data of the generative AI model, improving the accuracy of the generative AI model and making future analysis and generation more accurate.

[0772] In this way, the system automates a series of processes, from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, improving accuracy through feedback, and even optimizing wording by combining an emotion engine. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, improving work efficiency and risk management.

[0773] Example 2

[0774] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0775] There is a demand for a system that can respond to legal issues quickly and appropriately, even when employees are not familiar with legal matters. It is also necessary to optimize the content of documents and emails by reflecting the user's feelings, and ensure smooth correspondence between the parties involved. Conventional systems do not automate the identification of laws and regulations or risk countermeasures, which means that legal responses take time, and the system does not reflect the user's feelings, making it difficult to communicate appropriately.

[0776] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0777] In this invention, the server includes a terminal means for inputting business content, a server means for receiving the input business content, a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, a generative AI model means for generating countermeasures to address the identified risks, an automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, an automatic sending means for automatically sending the generated documents and emails to relevant parties, a feedback input means for allowing a user to input feedback on the content of the automatically generated documents and emails and reflecting this feedback in improving the accuracy of the generative AI model, an emotion engine for recognizing a user's emotions based on the user's input content and feedback, a wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine, and an emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means. This enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and documents and emails that reflect the user's emotions can be generated and sent, thereby facilitating communication.

[0778] The "terminal means for inputting business content" is a device equipped with an input device and an input form for a user to input business content in text format.

[0779] The "server means for receiving the input business content" is a server device for receiving the business content input by the user through the terminal means.

[0780] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to means that uses natural language processing technology to analyze input business content and identify relevant laws and regulations and potential risks.

[0781] A "generative AI model means for generating countermeasures for identified risks" is a means for using an artificial intelligence model to generate appropriate countermeasures for identified risks.

[0782] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a means for automatically creating necessary documents and e-mails based on the generated solutions.

[0783] The "automatic sending means for automatically sending the generated document or e-mail to the relevant person" is a means for automatically sending the generated document or e-mail to the designated relevant person.

[0784] "Feedback input means that allows users to input feedback on the content of automatically generated documents or emails and reflect it in improving the accuracy of the generative AI model" refers to a means that allows users to input opinions and suggestions for improvement on the content of automatically generated documents or emails, and uses that feedback information as learning data for the generative AI model.

[0785] "Emotion engine that recognizes user emotions based on user input and feedback" is an engine that analyzes and recognizes emotions from the content and feedback entered by the user.

[0786] The "wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine" is a means for optimally adjusting the content of generated documents and emails based on the user's emotions recognized by the emotion engine.

[0787] The "emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means" is a means for incorporating the emotion data recognized by the emotion engine into the learning data of the generative AI model.

[0788] The "template selection means" is a means for selecting an appropriate template for a document or email based on the identified risks and countermeasures.

[0789] The system of the present invention is designed to process legal work quickly and appropriately, and enables the generation of documents and emails that reflect the user's feelings. A specific embodiment of the present invention will be described below.

[0790] 1. System Configuration

[0791] The system consists of the following elements:

[0792] Terminal means for inputting business details

[0793] Server means for receiving input business details

[0794] Natural language processing means to identify relevant laws and potential risks from received business content

[0795] A generative AI model that generates countermeasures to address identified risks

[0796] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0797] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0798] A feedback input method that allows users to input feedback on the content of automatically generated documents and emails, and reflects this feedback in improving the accuracy of the generative AI model.

[0799] An emotion engine that recognizes user emotions based on user input and feedback

[0800] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[0801] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[0802] 2. Hardware and Software Used

[0803] Terminal means: The user uses an input device such as a computer or tablet to enter the details of the work in text format into a dedicated input form.

[0804] Server means: The server is a platform for receiving and processing business content sent from the terminal. The server may also use cloud-based services.

[0805] Natural language processing: For natural language processing, we use NLP libraries such as SpaCy and BERT to analyze the input business content and identify relevant laws and regulations and potential risks.

[0806] Generative AI model means: To generate countermeasures, we use generative AI models such as GPT-3 and BERT, which automatically generate appropriate countermeasures.

[0807] Automatic document generation method: To automatically generate documents and emails, a specified template is used. The template is automatically selected according to the business content and the solution.

[0808] Emotion Engine: For emotion recognition, emotion analysis algorithms (e.g., TextBlob, VADER) are used to identify emotions from user input and feedback.

[0809] 3. Specific Examples

[0810] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[0811] 1. Using the terminal means, the user enters "Plan a marketing strategy for new product A" into the input form and presses the send button.

[0812] 2. The terminal sends the entered business details to the server.

[0813] 3. The server analyzes the received data using natural language processing tools (e.g., SpaCy) to identify consumer protection laws and advertising regulations.

[0814] 4. The server sends a prompt to the generative AI model saying, "Please generate solutions to address the risk of violating advertising regulations," and receives the solutions.

[0815] 5. The server automatically generates a marketing plan using a template based on the solution.

[0816] 6. The server automatically sends the generated plan to the marketing and legal departments.

[0817] 7. The user checks the plan and enters feedback such as "Please correct this part."

[0818] 8. The server analyzes the feedback, recognizes the emotion of "urgency," and adjusts the tone of the document using language optimization techniques.

[0819] 9. The server reflects the recognized emotion data in the learning data of the generative AI model, aiming to improve accuracy from the next time onwards.

[0820] Example prompt sentence:

[0821] "Identify the laws and regulations and potential risks that need to be considered when developing a marketing strategy for new product A, and generate appropriate countermeasures."

[0822] "Generate appropriate documentation to comply with consumer protection laws and advertising regulations when developing your marketing strategies."

[0823] In this way, the present invention enables even users who are not familiar with legal matters to respond to legal matters quickly and appropriately, and facilitates smooth communication through the creation and transmission of documents and emails that reflect emotions.

[0824] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0825] Step 1:

[0826] The user uses the terminal means to input the details of the job in text format into the input form. For example, the user might input "Plan a marketing strategy for new product A" and press the send button.

[0827] Input: Job description (e.g., "Plan a marketing strategy for new product A")

[0828] Output: Text data of the work content is sent from the terminal

[0829] Step 2:

[0830] The terminal transmits the text data of the entered business details to the server.

[0831] Input: Text data of business content

[0832] Output: Data sent to the server

[0833] Step 3:

[0834] The server analyzes the received text data of the business content using natural language processing tools (e.g., SpaCy). As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" may be identified.

[0835] Input: Text data of business content

[0836] Data processing: Applying natural language processing to identify relevant laws and risks

[0837] Output: List of relevant laws and regulations and potential risks

[0838] Step 4:

[0839] The server requests the generative AI model to generate a solution based on the identified laws and risks. For example, it sends a prompt saying, "Please generate a solution to address the risk of violating advertising display regulations." The generative AI model generates an appropriate solution and sends it back to the server.

[0840] Input: List of relevant laws and regulations and potential risks, prompt text

[0841] Data computation: Using generative AI models to generate solutions

[0842] Output: A list of appropriate actions

[0843] Step 5:

[0844] Based on the received solutions, the server uses an automatic document generation means to automatically generate the necessary documents and emails. At this time, a template appropriate for the business content and solutions is selected, and specific documents and emails are created. For example, a "marketing plan" or "compliance checklist" is automatically generated.

[0845] Input: Solution list, template

[0846] Data processing: Generate documents and emails based on solutions and templates

[0847] Output: Generated documents and emails

[0848] Step 6:

[0849] The server automatically sends the generated documents and emails to the appropriate parties, including the user, relevant internal departments (e.g., marketing, legal), and external legal counsel.

[0850] Input: Generated documents and emails

[0851] Data processing: Send using email transmission protocols (e.g., SMTP)

[0852] Output: Documents and emails sent to stakeholders

[0853] Step 7:

[0854] The user checks the content of the automatically generated document or email and enters feedback as needed. For example, the user may enter feedback such as "Please correct this part." The feedback is sent from the terminal to the server.

[0855] Input: Feedback

[0856] Output: Feedback is sent to the server

[0857] Step 8:

[0858] The server receives the feedback and uses an emotion engine to analyze the user's emotions. For example, the emotion representing "urgency" is recognized. Based on this emotion, a wording optimizer adjusts the tone and wording of the generated document or email.

[0859] Input: Feedback

[0860] Data Computation: Emotion Analysis with Emotion Engine

[0861] Output: Recognized sentiment, optimized wording

[0862] Step 9:

[0863] The server reflects the emotion data recognized by the emotion engine in the learning data of the generative AI model, enabling more appropriate responses in subsequent analyses and document generation.

[0864] Input: Emotion data

[0865] Data processing: Reflecting emotion data in the training data of the generative AI model

[0866] Output: Updated training data for the generative AI model

[0867] In this way, a system is realized in which the processing at each step is linked, allowing even users unfamiliar with legal matters to respond quickly and appropriately to legal matters.

[0868] (Application example 2)

[0869] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0870] With conventional systems, it was difficult for employees without legal expertise to generate documents that complied with regulations, which increased the likelihood of mistakes and risks. Furthermore, the system was unable to respond flexibly to user sentiment, resulting in reduced operational efficiency. These issues must be resolved, particularly for online shopping sites, where compliance with regulations and communication based on user sentiment are crucial.

[0871] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes server means for receiving input business content, natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, generation AI model means for generating countermeasures for addressing the identified risks, automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, automatic sending means for automatically sending the generated documents and emails to relevant parties, emotion analysis means for recognizing emotions from user input content and feedback, and wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means. This enables appropriate responses to laws and regulations and flexible document generation based on user emotions.

[0872] The "terminal means for inputting business details" is a device equipped with an interface for a user to input business details in text format.

[0873] The "server means for receiving input business content" is a server for receiving and storing data sent by a user from a terminal means.

[0874] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to natural language processing technology for analyzing received text data and identifying relevant laws and regulations and potential risks.

[0875] A "generative AI model means for generating countermeasures to address identified risks" is an artificial intelligence model that generates methods and guidelines for addressing identified risks.

[0876] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[0877] "Automatic sending means for automatically sending generated documents and e-mails to relevant parties" refers to a system that has the function of automatically sending generated documents and e-mails to designated relevant parties.

[0878] "Emotion analysis means for recognizing emotions from user input and feedback" is a technology for determining emotions from text data and feedback entered by the user.

[0879] The "wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means" is a system that adjusts the tone and expression of generated documents and emails based on the emotions recognized by the emotion analysis means.

[0880] The "template selection means for selecting a document or email template based on identified risks and countermeasures" is a system that has the function of automatically selecting the most appropriate template based on risks and countermeasures.

[0881] "Feedback input means that allows users to input feedback on the contents of automatically generated documents and emails, and reflect this in improving the accuracy of the generative AI model" refers to a system that has the function of allowing users to provide feedback on automatically generated documents and emails, and to improve the generative AI model based on that feedback.

[0882] "A means for updating a generative AI model that reflects feedback content and emotional data based on generated documents and emails in the learning data of the generative AI model" is a technology that uses provided feedback and emotional data to update a generative AI model and improve the accuracy and flexibility of the model.

[0883] This embodiment of the present invention relates to a compliance assistance system for online shopping sites. The purpose of this system is to quickly and accurately perform legal checks when users release new products or services. A specific embodiment of this system will be described below.

[0884] System configuration

[0885] This system mainly consists of the following hardware and software:

[0886] Hardware: Smartphone

[0887] Software: Python, natural language processing libraries (spaCy, NLTK), sentiment analysis library (TextBlob), generative AI model (OpenAI GPT-3), Flask, REST API, Jinja2, smtplib

[0888] Program processing

[0889] 1. Input form: Using a smartphone app, users enter text descriptions of new products and services. This input form is built using HTML and JavaScript.

[0890] 2. Data reception: The data entered by the user is sent to the server via a REST API built using Flask. The server receives and stores this data.

[0891] 3. Natural Language Analysis: The server analyzes the received input data using Python and natural language processing libraries (spaCy, NLTK). This analysis identifies relevant laws and regulations and potential risks.

[0892] 4. Generative AI model: Based on the identified risks, a generative AI model (GPT-3) is used to generate countermeasures. The generative AI model is invoked using Python and the OpenAI API.

[0893] 5. Sentiment Analysis: TextBlob is used to analyze sentiment from user text input and feedback. This sentiment data is used to generate documents and emails.

[0894] 6. Automatic document generation: Based on the generated responses and the results of sentiment analysis, the necessary documents and emails are automatically generated using Jinja2.

[0895] 7. Automatic sending: Generated documents and emails are automatically sent to the relevant parties using Python's smtplib library.

[0896] 8. Feedback collection: Users can input feedback on the content of generated documents and emails through a smartphone app. This feedback is sent back to the server and reflected in the training data for the generative AI model.

[0897] In this way, appropriate measures for legal regulations and flexible document generation based on user feelings are realized.

[0898] Specific examples

[0899] For example, the user enters the following prompt text:

[0900] Develop a marketing strategy for new product A, including how to effectively appeal to the target market while complying with consumer protection laws and advertising regulations.

[0901] The server that receives this input uses natural language processing means to identify relevant laws and regulations, such as "consumer protection laws" and "regulations on advertising display," as well as risks. Then, the generative AI model means generates countermeasures for these risks. Based on the generated countermeasures, the automatic document generation means creates appropriate marketing plans and compliance checklists.

[0902] The sentiment analysis means analyzes the emotion of "effectively appealing" that the user emphasized in their input, and the wording optimization means optimizes the tone and expression of the document. Finally, the generated document or email is automatically sent to the relevant department or person in charge. The user can review the generated document and provide feedback as needed, which will improve the accuracy and responsiveness of the generation AI model from the next time onwards. In this way, business efficiency and risk management are improved.

[0903] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0904] Step 1:

[0905] User enters business details

[0906] A user enters a text description of a new product or service into an input form on a smartphone app. For example, they might enter, "I'm planning a marketing strategy for new product A. I want to include ways to effectively appeal to the target market while also complying with consumer protection laws and advertising regulations." This input data is sent from an input form built using HTML and JavaScript.

[0907] Step 2:

[0908] Data Receipt and Storage

[0909] Data on work details sent from the smartphone app is sent to the server via a REST API using Flask. The server receives this data and stores it in a database. An example of input data is text such as "Plan a marketing strategy for new product A."

[0910] Step 3:

[0911] natural language analysis

[0912] The server analyzes the received business text using natural language processing libraries (spaCy, NLTK). This process identifies relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display" are identified. The input data is text, and the output data is the name of the law and risk information.

[0913] Step 4:

[0914] Generative AI model generates solutions

[0915] The server uses a generative AI model (GPT-3) to generate appropriate countermeasures based on the identified risks. The generative AI model operates through Python and the OpenAI API, receiving legal names and risk information as input data. The output data is specific countermeasures and guidelines.

[0916] Step 5:

[0917] Emotion analysis

[0918] The server uses TextBlob to analyze the sentiment from the user's input text and feedback. This sentiment analysis extracts the user's emphasis and sentiment. The input data is text, and the output data is sentiment information such as positive, negative, and urgency.

[0919] Step 6:

[0920] Auto-generated documents

[0921] The server uses Jinja2 to automatically generate the necessary documents and emails based on the generated responses and the results of sentiment analysis. The input data are responses and sentiment data, and the output data is customized documents and emails based on templates. For example, a marketing plan or legal checklist based on regulations is generated.

[0922] Step 7:

[0923] Automatic transmission

[0924] The created documents and emails are automatically sent to the relevant parties using the Python smtplib library. The input data is the created document or email, and the output data is recorded on the server as a notification of the completion of sending.

[0925] Step 8:

[0926] Feedback collection

[0927] Users input feedback on generated documents and emails through a smartphone app. The feedback is sent to the server and stored in the database again. This data is used as training data for the generative AI model, helping to improve the accuracy of future document generation. The input data is the feedback content, and the output data is the improved response of the generative AI model.

[0928] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0929] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0930] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0931] [Third embodiment]

[0932] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0933] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0934] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0935] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0936] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0937] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0938] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0939] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0940] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0941] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0942] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0943] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0944] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[0945] System Overview

[0946] The system mainly consists of the following elements:

[0947] Terminal means for inputting business details

[0948] Server means for receiving input business details

[0949] Natural language processing means to identify relevant laws and potential risks from received business content

[0950] A generative AI model that generates countermeasures to address identified risks

[0951] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[0952] Automatic sending method for automatically sending generated documents and emails to relevant parties

[0953] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[0954] Example of a system

[0955] Terminal means

[0956] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[0957] Server Means

[0958] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[0959] Natural language processing and generative AI modeling tools

[0960] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[0961] Automatic document generation method

[0962] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[0963] Automatic transmission method

[0964] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[0965] Feedback Input Method

[0966] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[0967] Specific examples

[0968] For example, if a user enters "Plan a marketing strategy for new product A," the system will automatically go through the following steps to identify relevant laws and regulations, assess risks, present countermeasures, and generate and send documents.

[0969] 1. The user enters the details of the job into the terminal and sends it to the server.

[0970] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[0971] 3. The server uses a generative AI model to generate countermeasures for the identified risks.

[0972] 4. The server automatically generates appropriate documents and emails based on the corrective action.

[0973] 5. The server automatically sends the generated document to the marketing and legal departments.

[0974] 6. The user reviews the document on their device and provides feedback if necessary.

[0975] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[0976] The processing flow will be explained below.

[0977] Step 1:

[0978] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[0979] Step 2:

[0980] The device sends the entered business details to the server, and the data is securely transferred using an API.

[0981] Step 3:

[0982] The server receives the submitted business content and passes the received data to a natural language processing (NLP) engine for analysis.

[0983] Step 4:

[0984] The server uses a natural language processing engine to analyze the received business content, understand the structure and meaning of the sentences, and identify relevant laws and regulations and potential risks.

[0985] Step 5:

[0986] The server identifies relevant laws and regulations and potential risks. For example, it identifies "consumer protection laws" and "regulations on advertising display" and recognizes the risk of violations.

[0987] Step 6:

[0988] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[0989] Step 7:

[0990] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[0991] Step 8:

[0992] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[0993] Step 9:

[0994] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[0995] Step 10:

[0996] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[0997] Step 11:

[0998] The user checks the contents of documents generated on the device and emails sent, and confirms that the contents are appropriate.

[0999] Step 12:

[1000] Users can input feedback as needed, which is sent from the device to the server and used to improve the accuracy of the generative AI model.

[1001] In this way, the system automates everything from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, and improving accuracy through feedback. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, resulting in improved work efficiency and risk management.

[1002] Example 1

[1003] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1004] In the past, it was difficult for employees with insufficient legal knowledge to respond quickly and appropriately to legal issues. Furthermore, the entire process of identifying relevant laws and potential risks, developing countermeasures, and creating and sending documents required a great deal of time and effort. Furthermore, feedback on the content of generated documents and emails was not properly reflected, making it difficult to improve the accuracy of the system. This resulted in reduced work efficiency and inadequate risk management.

[1005] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1006] In this invention, the server includes a means for receiving business content, a means for identifying relevant laws and regulations and potential risks from the received business content, and a generative AI model means for generating countermeasures to address the identified risks. This allows even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters. Furthermore, by including a means for automatically generating necessary documents and emails based on the generated countermeasures, a means for automatically sending the generated documents and emails to relevant parties, and a means for users to input feedback on the content of the automatically generated documents and emails and reflect this feedback in improving the accuracy of the generative AI model, it is possible to further improve business efficiency and the accuracy of risk management.

[1007] "Business content" refers to text information related to a business that a user inputs into a terminal.

[1008] "Terminal means" refers to a device or its interface that allows a user to input business details.

[1009] The "server means" is a server that receives the business contents sent from the terminal and performs the processing.

[1010] "Natural language processing means" refers to technologies and algorithms used to analyze received business content and identify relevant laws and regulations and potential risks.

[1011] "Generative AI model means" refers to an artificial intelligence model for generating countermeasures to identified risks.

[1012] The "automatic document generation means" is a system or program for automatically creating the necessary documents or emails based on the generated solutions.

[1013] "Automatic sending means" refers to a system or method for automatically sending generated documents or emails to relevant parties.

[1014] "Feedback input means" refers to a system or method that allows users to input feedback on the contents of automatically generated documents or emails, and uses that feedback to improve the accuracy of the generative AI model.

[1015] "Template selector" means a system or algorithm for selecting document or email templates based on identified risks and countermeasures.

[1016] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[1017] System Configuration

[1018] The system includes the following major hardware and software elements:

[1019] Terminal means: A device for inputting work details. For example, a PC or tablet. The user inputs the work details and sends them to the server.

[1020] Server means: A server that receives and processes the business content sent from the terminal. Specifically, it runs natural language processing (NLP) libraries (e.g., SpaCy, BERT) and generative AI models (e.g., GPT-3).

[1021] Natural language processing means: A program that runs on the server and analyzes the received business content to identify relevant laws and regulations and potential risks.

[1022] Generative AI model means: An artificial intelligence model integrated into the server generates countermeasures for identified risks.

[1023] Automatic document generation: Automatically create documents and emails based on the solutions generated on the server. A template engine (e.g., Jinja2) is used.

[1024] Automatic sending means: A system that sends generated documents and emails to the relevant parties, for example, via an SMTP server.

[1025] Feedback input method: The user checks the contents of the generated documents and emails, and inputs feedback as necessary to improve the accuracy of the generative AI model. The feedback data is stored in a database (e.g., PostgreSQL, MySQL).

[1026] Template selection tools: Tools for selecting document and email templates based on identified risks and treatments.

[1027] Example of a system

[1028] Terminal means

[1029] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[1030] Server Means

[1031] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[1032] Natural language processing and generative AI modeling tools

[1033] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[1034] Automatic document generation method

[1035] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[1036] Automatic transmission method

[1037] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[1038] Feedback Input Method

[1039] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[1040] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[1041] Specific examples

[1042] For example, if a user inputs "Plan a marketing strategy for new product A," the process will go through the following steps:

[1043] 1. The business details entered by the user are sent to the server from the terminal.

[1044] 2. The server receives the business details and performs natural language processing to identify relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display."

[1045] 3. The server uses the generative AI model to generate a solution, such as suggesting "create advertising copy that complies with the Consumer Protection Act."

[1046] 4. The server automatically generates documents and emails based on the measures taken, such as a "marketing plan" or a "compliance checklist."

[1047] 5. The server automatically sends the generated document to the relevant parties.

[1048] 6. The user reviews the document and provides feedback if necessary.

[1049] Examples of prompts

[1050] An example of a prompt to input to a generative AI model is: "Analyze the laws and potential risks related to the marketing strategy for new product A, and propose appropriate countermeasures."

[1051] Based on this prompt, the generative AI model suggests appropriate actions, and the system then generates the necessary documents and emails.

[1052] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1053] Step 1:

[1054] The user inputs the details of the job into the terminal and sends it to the server.

[1055] Input: Enter the job description in text format as "Plan a marketing strategy for new product A."

[1056] Output: Text data of the work content sent from the terminal to the server.

[1057] Specific operation: The user enters "Plan a marketing strategy for new product A" into the input form on the terminal and clicks the submit button.

[1058] Step 2:

[1059] The server receives the business content and analyzes it using natural language processing means.

[1060] Input: Text data of the work content sent from the terminal.

[1061] Output: Data on relevant laws and regulations and potential risks extracted from business operations.

[1062] Specific operation: The server analyzes the text received from the device, "Plan a marketing strategy for new product A," using a natural language processing tool (such as SpaCy or BERT), and identifies relevant laws and regulations such as "consumer protection laws" and "regulations on advertising display," as well as potential risks.

[1063] Step 3:

[1064] The server uses the generative AI model to generate countermeasures for the identified risks.

[1065] Input: Data on relevant laws and regulations and potential risks identified through natural language processing.

[1066] Output: Text data of relevant laws and regulations and countermeasures for potential risks.

[1067] Specific operation: When a risk related to the Consumer Protection Act is identified, the server sends a prompt to the generative AI model (e.g., GPT-3) saying, "Please create advertising text that complies with the Consumer Protection Act," and generates a solution (appropriate advertising text and display method).

[1068] Step 4:

[1069] The server automatically generates the necessary documents and emails based on the generated solutions.

[1070] Input: Text data of solutions obtained from the generative AI model.

[1071] Output: Auto-generated documents and email data.

[1072] Specific operation: The server receives the generated solutions in text format and uses a template engine (e.g., Jinja2) to automatically generate a "marketing plan in accordance with consumer protection laws" and a "compliance checklist."

[1073] Step 5:

[1074] The server automatically sends generated documents and emails to the relevant parties.

[1075] Input: Data from automatically generated documents and emails.

[1076] Output: The sent document or email reaches the relevant person.

[1077] Specific operation: The server sends the generated "marketing plan" via email to the user, marketing department, and legal department via the SMTP server.

[1078] Step 6:

[1079] The user checks the generated documents and emails and provides feedback.

[1080] Input: The contents of the document or email sent.

[1081] Output: Feedback data from users.

[1082] Specific actions: The user reviews the received "Marketing Plan" and submits feedback by entering a comment such as "This ad copy is appropriate, but I would like it to be a little more specific."

[1083] Step 7:

[1084] The server receives user feedback and uses it to improve the accuracy of the generative AI model.

[1085] Input: Feedback data submitted by the user.

[1086] Output: Updated generative AI model data.

[1087] Specific operation: The server stores the received feedback in a database and uses the feedback data as training data for the generative AI model to improve the accuracy of the model.

[1088] (Application example 1)

[1089] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1090] Conventional regulatory compliance and risk management systems have had difficulty identifying work content and providing appropriate countermeasures quickly and accurately. In particular, with regard to work robots in factories, immediate responses to complex regulations and risks are required, but if on-site personnel lack sufficient legal knowledge, responses may be delayed. Therefore, there is a need for systems that improve the efficiency of regulatory compliance and risk management, thereby improving productivity and safety.

[1091] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1092] In this invention, the server includes a generation AI model means for providing appropriate countermeasures related to received laws and regulations and risks, a terminal means for inputting business content, a server means for receiving the input business content, and a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, thereby enabling prompt and appropriate responses to laws and regulations and risks.

[1093] The "terminal means for inputting work content" refers to a device that allows workers and managers in a factory to input work content and work procedures in voice or text format.

[1094] The "server means for receiving input business content" is a central processing unit for receiving data from the terminal into which the business content has been input, and for managing and processing the data.

[1095] "Natural language processing means for identifying relevant laws and potential risks from received business content" refers to a system that includes natural language processing (NLP) technology for analyzing the text data of received business content and identifying relevant laws and risks.

[1096] "Generative AI model means for generating countermeasures for identified risks" refers to an artificial intelligence (AI) model for generating appropriate countermeasures and guidelines for identified laws and regulations and risks.

[1097] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[1098] "Automatic sending means for automatically sending generated documents and emails to relevant parties" refers to a device or system for automatically sending generated documents and emails to appropriate relevant parties (e.g., factory managers or legal departments).

[1099] "Generative AI model means for providing appropriate countermeasures related to received laws and regulations and risks" is a generative AI model for providing preventive measures and countermeasures based on received laws and regulations and risks.

[1100] The "template selection means" is a system that has the function of selecting an appropriate template according to the identified risks and countermeasures.

[1101] A "feedback input means" is a device or system that allows a user to input feedback on the content of automatically generated documents or emails, and that allows that feedback to be reflected in improving the generative AI model.

[1102] The present invention provides a system for automating a series of processes for compliance with regulations and risk management, from identifying the work content of work robots in a factory to presenting appropriate countermeasures, automatically generating and sending documents, and collecting feedback. A detailed embodiment of this system will be described.

[1103] System configuration

[1104] The system consists of the following hardware and software means:

[1105] 1. Terminal means: A device used to input work content. Input can be done by voice or text. A specific example is the microphone and keyboard installed on a work robot in a factory.

[1106] 2. Server means: A central processing unit that receives and processes input business content. It has a built-in database and analytical model, and performs high-speed and accurate processing.

[1107] 3. Natural Language Processing (NLP): Analyzes business content and identifies relevant laws and regulations and potential risks. For example, a natural language processing library such as Spacy is used.

[1108] 4. Generative AI model means: Generate appropriate countermeasures to address identified risks. For example, the GPT-3 model using the OpenAI API.

[1109] 5. Automatic document generation means: This has the function of automatically creating the necessary documents and emails based on the generated countermeasures. A template-based document generation system is used.

[1110] 6. Automated Delivery: Automatically send generated documents and emails to relevant parties. This includes email systems and internal notification systems.

[1111] 7. Feedback input means: A system in which users can input feedback on the contents of automatically generated documents and emails, and that feedback is reflected in improving the generative AI model.

[1112] Program processing explanation

[1113] The server receives the business content from the terminal and analyzes the input content using natural language processing means. From the analyzed content, relevant laws and regulations and potential risks are identified, and based on that, a generative AI model provides countermeasures. Next, an automatic document generation means automatically generates documents and emails based on the countermeasures, which are then sent to relevant parties using an automatic sending means. The entire process aims to improve business efficiency and ensure swift and appropriate compliance with laws and regulations and risk management.

[1114] Examples of hardware and software used include:

[1115] Hardware: Factory robots, servers, smart devices for administrators

[1116] Software: Spacy (natural language processing library), OpenAI GPT-3 API (generative AI model), email system, template-based document generation system

[1117] Adding specific examples

[1118] For example, when introducing a new product assembly line in a factory, a manager inputs "Introducing a new product assembly line" into the robot's voice input system. The server receives this information and uses natural language processing to identify relevant laws and regulations, such as the Industrial Safety and Health Act, and potential risks, such as "Worker Safety Risks" and "Environmental Pollution Risks." The generative AI model then generates countermeasures based on the following prompt:

[1119] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[1120] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[1121] Please tell me the appropriate countermeasures.

[1122] Based on the proposed measures, the server automatically generates documents such as checklists and guidelines and sends them to the factory manager and legal department. The manager reviews the received documents and provides feedback as necessary, which the system uses to improve the system in the future.

[1123] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[1124] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1125] Step 1:

[1126] The user inputs the job details. A worker or manager in the factory inputs the specific job details (for example, "Introduce an assembly line for a new product") into the work robot's terminal using voice or text format. The input data is saved on the terminal and sent to the server.

[1127] Input: Task details (e.g., "Install an assembly line for a new product")

[1128] Output: Text data transmitted to the terminal, data sent to the server

[1129] Step 2:

[1130] The server receives the work content. The entered work content data is received and prepared for analysis. The received work content is saved to proceed to the next analysis stage.

[1131] Input: Text data sent from the terminal

[1132] Output: Data ready for analysis

[1133] Step 3:

[1134] The business content is analyzed using natural language processing. The server analyzes the received business content using a natural language processing library such as Spacy, and identifies relevant laws and potential risks. For example, relevant laws such as the "Occupational Safety and Health Act" and "Environmental Protection Act" and risks such as "worker safety risks" and "environmental pollution risks" are identified.

[1135] Input: Data ready for analysis

[1136] Output: Identified regulations and potential risks

[1137] Step 4:

[1138] Countermeasures are generated using a generative AI model. The server uses a generative AI model such as OpenAI GPT-3 to generate appropriate countermeasures for the identified laws and potential risks. For this, the following prompt sentences are used:

[1139] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[1140] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[1141] Please tell me the appropriate countermeasures.

[1142] Input: Identified laws and regulations and potential risks, prompt text

[1143] Output: Generated countermeasures

[1144] Step 5:

[1145] Documents and emails are automatically generated using an automatic document generation means. Based on the generated countermeasures, the server uses a template-based automatic document generation system to create the necessary documents and emails. For example, "checklists" and "guidelines" are created.

[1146] Input: Generated countermeasures

[1147] Output: Generated documents and emails

[1148] Step 6:

[1149] Send documents and emails to relevant parties using automated methods. The server automatically sends generated documents and emails to relevant parties, such as factory management or the legal department. This process uses email systems and internal notification systems.

[1150] Input: Generated documents and emails

[1151] Output: Documents and emails sent to stakeholders

[1152] Step 7:

[1153] The user enters feedback. Administrators and other relevant parties review the contents of received documents and emails and enter feedback as necessary. This feedback is sent to the server and used to improve the accuracy of the generative AI model.

[1154] Input: User feedback

[1155] Output: Feedback data stored on the server

[1156] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[1157] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1158] The system of the present invention is designed to enable even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and by adding a function to recognize and respond to the user's emotions, the content of documents and emails can be optimized. The following describes in detail the embodiments of the present invention.

[1159] System Overview

[1160] The system mainly consists of the following elements:

[1161] Terminal means for inputting business details

[1162] Server means for receiving input business details

[1163] Natural language processing means to identify relevant laws and potential risks from received business content

[1164] A generative AI model that generates countermeasures to address identified risks

[1165] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[1166] Automatic sending method for automatically sending generated documents and emails to relevant parties

[1167] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[1168] An emotion engine that recognizes emotions based on user input and feedback

[1169] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[1170] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[1171] Example of a system

[1172] Terminal means

[1173] The user inputs the details of the task in text format using the input form on the terminal. For example, the user may input "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[1174] Server Means

[1175] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[1176] Natural language processing and generative AI modeling tools

[1177] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[1178] Automatic document generation method

[1179] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[1180] Automatic transmission method

[1181] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[1182] Feedback Input Method

[1183] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[1184] Emotion engine and wording optimization

[1185] The emotion engine analyzes user input and feedback and recognizes the emotion. For example, if a user inputs an urgent task, the emotion engine recognizes this as the emotion representing "urgent." The recognized emotion is used by the wording optimization tool to optimize the tone and wording of generated documents and emails. This allows recipients to better understand the content and respond quickly.

[1186] Emotion data reflection method

[1187] The emotion data recognized by the emotion engine is also reflected in the learning data of the generative AI model, improving the model's accuracy and flexibility, and enabling more appropriate responses in future business content analysis and document generation.

[1188] Specific examples

[1189] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[1190] 1. The user enters the details of the job into the terminal and sends it to the server.

[1191] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[1192] 3. The server requests the generative AI model to generate risk countermeasures and receives the appropriate countermeasures.

[1193] 4. The server automatically generates documents and emails based on the corrective action.

[1194] 5. Language optimization tools adjust the tone of documents and emails based on user sentiment.

[1195] 6. The server automatically sends the generated document to the marketing and legal departments.

[1196] 7. The user checks the results on the device and provides feedback if necessary.

[1197] 8. The feedback content is sent to the server, where it is emotionally analyzed by the emotion engine and reflected in the training data of the generative AI model.

[1198] In this way, the present invention is highly effective in improving business efficiency and risk management by automating the generation and transmission of legal documents that reflect the user's feelings.

[1199] The processing flow will be explained below.

[1200] Step 1:

[1201] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[1202] Step 2:

[1203] The device sends the entered business details to the server using an API, which transfers the data securely.

[1204] Step 3:

[1205] The server receives the submitted business details and passes the received data to the natural language processing engine.

[1206] Step 4:

[1207] The server uses a natural language processing engine to analyze the received business content. As a result of the analysis, the structure and meaning of the sentence are understood, and relevant laws and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" are identified.

[1208] Step 5:

[1209] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[1210] Step 6:

[1211] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[1212] Step 7:

[1213] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[1214] Step 8:

[1215] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[1216] Step 9:

[1217] The user enters feedback from the terminal, providing additional information and suggestions for improvement based on the content of the generated documents and emails.

[1218] Step 10:

[1219] The server receives the feedback and passes it to the emotion engine for analysis. The user's emotion is identified from the feedback content. For example, the emotion meaning "urgency" is recognized.

[1220] Step 11:

[1221] The server optimizes the wording of documents and emails based on the emotional data from the emotion engine, adjusting the tone and wording as needed.

[1222] Step 12:

[1223] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[1224] Step 13:

[1225] The server reflects the emotion data identified by the emotion engine in the learning data of the generative AI model, improving the accuracy of the generative AI model and making future analysis and generation more accurate.

[1226] In this way, the system automates a series of processes, from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, improving accuracy through feedback, and even optimizing wording by combining an emotion engine. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, improving work efficiency and risk management.

[1227] Example 2

[1228] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1229] There is a demand for a system that can respond to legal issues quickly and appropriately, even when employees are not familiar with legal matters. It is also necessary to optimize the content of documents and emails by reflecting the user's feelings, and ensure smooth correspondence between the parties involved. Conventional systems do not automate the identification of laws and regulations or risk countermeasures, which means that legal responses take time, and the system does not reflect the user's feelings, making it difficult to communicate appropriately.

[1230] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1231] In this invention, the server includes a terminal means for inputting business content, a server means for receiving the input business content, a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, a generative AI model means for generating countermeasures to address the identified risks, an automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, an automatic sending means for automatically sending the generated documents and emails to relevant parties, a feedback input means for allowing a user to input feedback on the content of the automatically generated documents and emails and reflecting this feedback in improving the accuracy of the generative AI model, an emotion engine for recognizing a user's emotions based on the user's input content and feedback, a wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine, and an emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means. This enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and documents and emails that reflect the user's emotions can be generated and sent, thereby facilitating communication.

[1232] The "terminal means for inputting business content" is a device equipped with an input device and an input form for a user to input business content in text format.

[1233] The "server means for receiving the input business content" is a server device for receiving the business content input by the user through the terminal means.

[1234] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to means that uses natural language processing technology to analyze input business content and identify relevant laws and regulations and potential risks.

[1235] A "generative AI model means for generating countermeasures for identified risks" is a means for using an artificial intelligence model to generate appropriate countermeasures for identified risks.

[1236] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a means for automatically creating necessary documents and e-mails based on the generated solutions.

[1237] The "automatic sending means for automatically sending the generated document or e-mail to the relevant person" is a means for automatically sending the generated document or e-mail to the designated relevant person.

[1238] "Feedback input means that allows users to input feedback on the content of automatically generated documents or emails and reflect it in improving the accuracy of the generative AI model" refers to a means that allows users to input opinions and suggestions for improvement on the content of automatically generated documents or emails, and uses that feedback information as learning data for the generative AI model.

[1239] "Emotion engine that recognizes user emotions based on user input and feedback" is an engine that analyzes and recognizes emotions from the content and feedback entered by the user.

[1240] The "wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine" is a means for optimally adjusting the content of generated documents and emails based on the user's emotions recognized by the emotion engine.

[1241] The "emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means" is a means for incorporating the emotion data recognized by the emotion engine into the learning data of the generative AI model.

[1242] The "template selection means" is a means for selecting an appropriate template for a document or email based on the identified risks and countermeasures.

[1243] The system of the present invention is designed to process legal work quickly and appropriately, and enables the generation of documents and emails that reflect the user's feelings. A specific embodiment of the present invention will be described below.

[1244] 1. System Configuration

[1245] The system consists of the following elements:

[1246] Terminal means for inputting business details

[1247] Server means for receiving input business details

[1248] Natural language processing means to identify relevant laws and potential risks from received business content

[1249] A generative AI model that generates countermeasures to address identified risks

[1250] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[1251] Automatic sending method for automatically sending generated documents and emails to relevant parties

[1252] A feedback input method that allows users to input feedback on the content of automatically generated documents and emails, and reflects this feedback in improving the accuracy of the generative AI model.

[1253] An emotion engine that recognizes user emotions based on user input and feedback

[1254] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[1255] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[1256] 2. Hardware and Software Used

[1257] Terminal means: The user uses an input device such as a computer or tablet to enter the details of the work in text format into a dedicated input form.

[1258] Server means: The server is a platform for receiving and processing business content sent from the terminal. The server may also use cloud-based services.

[1259] Natural language processing: For natural language processing, we use NLP libraries such as SpaCy and BERT to analyze the input business content and identify relevant laws and regulations and potential risks.

[1260] Generative AI model means: To generate countermeasures, we use generative AI models such as GPT-3 and BERT, which automatically generate appropriate countermeasures.

[1261] Automatic document generation method: To automatically generate documents and emails, a specified template is used. The template is automatically selected according to the business content and the solution.

[1262] Emotion Engine: For emotion recognition, emotion analysis algorithms (e.g., TextBlob, VADER) are used to identify emotions from user input and feedback.

[1263] 3. Specific Examples

[1264] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[1265] 1. Using the terminal means, the user enters "Plan a marketing strategy for new product A" into the input form and presses the send button.

[1266] 2. The terminal sends the entered business details to the server.

[1267] 3. The server analyzes the received data using natural language processing tools (e.g., SpaCy) to identify consumer protection laws and advertising regulations.

[1268] 4. The server sends a prompt to the generative AI model saying, "Please generate solutions to address the risk of violating advertising regulations," and receives the solutions.

[1269] 5. The server automatically generates a marketing plan using a template based on the solution.

[1270] 6. The server automatically sends the generated plan to the marketing and legal departments.

[1271] 7. The user checks the plan and enters feedback such as "Please correct this part."

[1272] 8. The server analyzes the feedback, recognizes the emotion of "urgency," and adjusts the tone of the document using language optimization techniques.

[1273] 9. The server reflects the recognized emotion data in the learning data of the generative AI model, aiming to improve accuracy from the next time onwards.

[1274] Example prompt sentence:

[1275] "Identify the laws and regulations and potential risks that need to be considered when developing a marketing strategy for new product A, and generate appropriate countermeasures."

[1276] "Generate appropriate documentation to comply with consumer protection laws and advertising regulations when developing your marketing strategies."

[1277] In this way, the present invention enables even users who are not familiar with legal matters to respond to legal matters quickly and appropriately, and facilitates smooth communication through the creation and transmission of documents and emails that reflect emotions.

[1278] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1279] Step 1:

[1280] The user uses the terminal means to input the details of the job in text format into the input form. For example, the user might input "Plan a marketing strategy for new product A" and press the send button.

[1281] Input: Job description (e.g., "Plan a marketing strategy for new product A")

[1282] Output: Text data of the work content is sent from the terminal

[1283] Step 2:

[1284] The terminal transmits the text data of the entered business details to the server.

[1285] Input: Text data of business content

[1286] Output: Data sent to the server

[1287] Step 3:

[1288] The server analyzes the received text data of the business content using natural language processing tools (e.g., SpaCy). As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" may be identified.

[1289] Input: Text data of business content

[1290] Data processing: Applying natural language processing to identify relevant laws and risks

[1291] Output: List of relevant laws and regulations and potential risks

[1292] Step 4:

[1293] The server requests the generative AI model to generate a solution based on the identified laws and risks. For example, it sends a prompt saying, "Please generate a solution to address the risk of violating advertising display regulations." The generative AI model generates an appropriate solution and sends it back to the server.

[1294] Input: List of relevant laws and regulations and potential risks, prompt text

[1295] Data computation: Using generative AI models to generate solutions

[1296] Output: A list of appropriate actions

[1297] Step 5:

[1298] Based on the received solutions, the server uses an automatic document generation means to automatically generate the necessary documents and emails. At this time, a template appropriate for the business content and solutions is selected, and specific documents and emails are created. For example, a "marketing plan" or "compliance checklist" is automatically generated.

[1299] Input: Solution list, template

[1300] Data processing: Generate documents and emails based on solutions and templates

[1301] Output: Generated documents and emails

[1302] Step 6:

[1303] The server automatically sends the generated documents and emails to the appropriate parties, including the user, relevant internal departments (e.g., marketing, legal), and external legal counsel.

[1304] Input: Generated documents and emails

[1305] Data processing: Send using email transmission protocols (e.g., SMTP)

[1306] Output: Documents and emails sent to stakeholders

[1307] Step 7:

[1308] The user checks the content of the automatically generated document or email and enters feedback as needed. For example, the user may enter feedback such as "Please correct this part." The feedback is sent from the terminal to the server.

[1309] Input: Feedback

[1310] Output: Feedback is sent to the server

[1311] Step 8:

[1312] The server receives the feedback and uses an emotion engine to analyze the user's emotions. For example, the emotion representing "urgency" is recognized. Based on this emotion, a wording optimizer adjusts the tone and wording of the generated document or email.

[1313] Input: Feedback

[1314] Data Computation: Emotion Analysis with Emotion Engine

[1315] Output: Recognized sentiment, optimized wording

[1316] Step 9:

[1317] The server reflects the emotion data recognized by the emotion engine in the learning data of the generative AI model, enabling more appropriate responses in subsequent analyses and document generation.

[1318] Input: Emotion data

[1319] Data processing: Reflecting emotion data in the training data of the generative AI model

[1320] Output: Updated training data for the generative AI model

[1321] In this way, a system is realized in which the processing at each step is linked, allowing even users unfamiliar with legal matters to respond quickly and appropriately to legal matters.

[1322] (Application example 2)

[1323] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1324] With conventional systems, it was difficult for employees without legal expertise to generate documents that complied with regulations, which increased the likelihood of mistakes and risks. Furthermore, the system was unable to respond flexibly to user sentiment, resulting in reduced operational efficiency. These issues must be resolved, particularly for online shopping sites, where compliance with regulations and communication based on user sentiment are crucial.

[1325] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes server means for receiving input business content, natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, generation AI model means for generating countermeasures for addressing the identified risks, automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, automatic sending means for automatically sending the generated documents and emails to relevant parties, emotion analysis means for recognizing emotions from user input content and feedback, and wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means. This enables appropriate responses to laws and regulations and flexible document generation based on user emotions.

[1326] The "terminal means for inputting business details" is a device equipped with an interface for a user to input business details in text format.

[1327] The "server means for receiving input business content" is a server for receiving and storing data sent by a user from a terminal means.

[1328] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to natural language processing technology for analyzing received text data and identifying relevant laws and regulations and potential risks.

[1329] A "generative AI model means for generating countermeasures to address identified risks" is an artificial intelligence model that generates methods and guidelines for addressing identified risks.

[1330] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[1331] "Automatic sending means for automatically sending generated documents and e-mails to relevant parties" refers to a system that has the function of automatically sending generated documents and e-mails to designated relevant parties.

[1332] "Emotion analysis means for recognizing emotions from user input and feedback" is a technology for determining emotions from text data and feedback entered by the user.

[1333] The "wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means" is a system that adjusts the tone and expression of generated documents and emails based on the emotions recognized by the emotion analysis means.

[1334] The "template selection means for selecting a document or email template based on identified risks and countermeasures" is a system that has the function of automatically selecting the most appropriate template based on risks and countermeasures.

[1335] "Feedback input means that allows users to input feedback on the contents of automatically generated documents and emails, and reflect this in improving the accuracy of the generative AI model" refers to a system that has the function of allowing users to provide feedback on automatically generated documents and emails, and to improve the generative AI model based on that feedback.

[1336] "A means for updating a generative AI model that reflects feedback content and emotional data based on generated documents and emails in the learning data of the generative AI model" is a technology that uses provided feedback and emotional data to update a generative AI model and improve the accuracy and flexibility of the model.

[1337] This embodiment of the present invention relates to a compliance assistance system for online shopping sites. The purpose of this system is to quickly and accurately perform legal checks when users release new products or services. A specific embodiment of this system will be described below.

[1338] System configuration

[1339] This system mainly consists of the following hardware and software:

[1340] Hardware: Smartphone

[1341] Software: Python, natural language processing libraries (spaCy, NLTK), sentiment analysis library (TextBlob), generative AI model (OpenAI GPT-3), Flask, REST API, Jinja2, smtplib

[1342] Program processing

[1343] 1. Input form: Using a smartphone app, users enter text descriptions of new products and services. This input form is built using HTML and JavaScript.

[1344] 2. Data reception: The data entered by the user is sent to the server via a REST API built using Flask. The server receives and stores this data.

[1345] 3. Natural Language Analysis: The server analyzes the received input data using Python and natural language processing libraries (spaCy, NLTK). This analysis identifies relevant laws and regulations and potential risks.

[1346] 4. Generative AI model: Based on the identified risks, a generative AI model (GPT-3) is used to generate countermeasures. The generative AI model is invoked using Python and the OpenAI API.

[1347] 5. Sentiment Analysis: TextBlob is used to analyze sentiment from user text input and feedback. This sentiment data is used to generate documents and emails.

[1348] 6. Automatic document generation: Based on the generated responses and the results of sentiment analysis, the necessary documents and emails are automatically generated using Jinja2.

[1349] 7. Automatic sending: Generated documents and emails are automatically sent to the relevant parties using Python's smtplib library.

[1350] 8. Feedback collection: Users can input feedback on the content of generated documents and emails through a smartphone app. This feedback is sent back to the server and reflected in the training data for the generative AI model.

[1351] In this way, appropriate measures for legal regulations and flexible document generation based on user feelings are realized.

[1352] Specific examples

[1353] For example, the user enters the following prompt text:

[1354] Develop a marketing strategy for new product A, including how to effectively appeal to the target market while complying with consumer protection laws and advertising regulations.

[1355] The server that receives this input uses natural language processing means to identify relevant laws and regulations, such as "consumer protection laws" and "regulations on advertising display," as well as risks. Then, the generative AI model means generates countermeasures for these risks. Based on the generated countermeasures, the automatic document generation means creates appropriate marketing plans and compliance checklists.

[1356] The sentiment analysis means analyzes the emotion of "effectively appealing" that the user emphasized in their input, and the wording optimization means optimizes the tone and expression of the document. Finally, the generated document or email is automatically sent to the relevant department or person in charge. The user can review the generated document and provide feedback as needed, which will improve the accuracy and responsiveness of the generation AI model from the next time onwards. In this way, business efficiency and risk management are improved.

[1357] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1358] Step 1:

[1359] User enters business details

[1360] A user enters a text description of a new product or service into an input form on a smartphone app. For example, they might enter, "I'm planning a marketing strategy for new product A. I want to include ways to effectively appeal to the target market while also complying with consumer protection laws and advertising regulations." This input data is sent from an input form built using HTML and JavaScript.

[1361] Step 2:

[1362] Data Receipt and Storage

[1363] Data on work details sent from the smartphone app is sent to the server via a REST API using Flask. The server receives this data and stores it in a database. An example of input data is text such as "Plan a marketing strategy for new product A."

[1364] Step 3:

[1365] natural language analysis

[1366] The server analyzes the received business text using natural language processing libraries (spaCy, NLTK). This process identifies relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display" are identified. The input data is text, and the output data is the name of the law and risk information.

[1367] Step 4:

[1368] Generative AI model generates solutions

[1369] The server uses a generative AI model (GPT-3) to generate appropriate countermeasures based on the identified risks. The generative AI model operates through Python and the OpenAI API, receiving legal names and risk information as input data. The output data is specific countermeasures and guidelines.

[1370] Step 5:

[1371] Emotion analysis

[1372] The server uses TextBlob to analyze the sentiment from the user's input text and feedback. This sentiment analysis extracts the user's emphasis and sentiment. The input data is text, and the output data is sentiment information such as positive, negative, and urgency.

[1373] Step 6:

[1374] Auto-generated documents

[1375] The server uses Jinja2 to automatically generate the necessary documents and emails based on the generated responses and the results of sentiment analysis. The input data are responses and sentiment data, and the output data is customized documents and emails based on templates. For example, a marketing plan or legal checklist based on regulations is generated.

[1376] Step 7:

[1377] Automatic transmission

[1378] The created documents and emails are automatically sent to the relevant parties using the Python smtplib library. The input data is the created document or email, and the output data is recorded on the server as a notification of the completion of sending.

[1379] Step 8:

[1380] Feedback collection

[1381] Users input feedback on generated documents and emails through a smartphone app. The feedback is sent to the server and stored in the database again. This data is used as training data for the generative AI model, helping to improve the accuracy of future document generation. The input data is the feedback content, and the output data is the improved response of the generative AI model.

[1382] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1383] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1384] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1385] [Fourth embodiment]

[1386] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1387] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1388] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1389] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1390] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1391] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1392] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1393] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1394] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1395] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1396] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1397] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1398] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1399] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[1400] System Overview

[1401] The system mainly consists of the following elements:

[1402] Terminal means for inputting business details

[1403] Server means for receiving input business details

[1404] Natural language processing means to identify relevant laws and potential risks from received business content

[1405] A generative AI model that generates countermeasures to address identified risks

[1406] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[1407] Automatic sending method for automatically sending generated documents and emails to relevant parties

[1408] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[1409] Example of a system

[1410] Terminal means

[1411] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[1412] Server Means

[1413] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[1414] Natural language processing and generative AI modeling tools

[1415] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[1416] Automatic document generation method

[1417] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[1418] Automatic transmission method

[1419] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[1420] Feedback Input Method

[1421] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[1422] Specific examples

[1423] For example, if a user enters "Plan a marketing strategy for new product A," the system will automatically go through the following steps to identify relevant laws and regulations, assess risks, present countermeasures, and generate and send documents.

[1424] 1. The user enters the details of the job into the terminal and sends it to the server.

[1425] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[1426] 3. The server uses a generative AI model to generate countermeasures for the identified risks.

[1427] 4. The server automatically generates appropriate documents and emails based on the corrective action.

[1428] 5. The server automatically sends the generated document to the marketing and legal departments.

[1429] 6. The user reviews the document on their device and provides feedback if necessary.

[1430] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[1431] The processing flow will be explained below.

[1432] Step 1:

[1433] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[1434] Step 2:

[1435] The device sends the entered business details to the server, and the data is securely transferred using an API.

[1436] Step 3:

[1437] The server receives the submitted business content and passes the received data to a natural language processing (NLP) engine for analysis.

[1438] Step 4:

[1439] The server uses a natural language processing engine to analyze the received business content, understand the structure and meaning of the sentences, and identify relevant laws and regulations and potential risks.

[1440] Step 5:

[1441] The server identifies relevant laws and regulations and potential risks. For example, it identifies "consumer protection laws" and "regulations on advertising display" and recognizes the risk of violations.

[1442] Step 6:

[1443] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[1444] Step 7:

[1445] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[1446] Step 8:

[1447] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[1448] Step 9:

[1449] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[1450] Step 10:

[1451] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[1452] Step 11:

[1453] The user checks the contents of documents generated on the device and emails sent, and confirms that the contents are appropriate.

[1454] Step 12:

[1455] Users can input feedback as needed, which is sent from the device to the server and used to improve the accuracy of the generative AI model.

[1456] In this way, the system automates everything from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, and improving accuracy through feedback. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, resulting in improved work efficiency and risk management.

[1457] Example 1

[1458] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1459] In the past, it was difficult for employees with insufficient legal knowledge to respond quickly and appropriately to legal issues. Furthermore, the entire process of identifying relevant laws and potential risks, developing countermeasures, and creating and sending documents required a great deal of time and effort. Furthermore, feedback on the content of generated documents and emails was not properly reflected, making it difficult to improve the accuracy of the system. This resulted in reduced work efficiency and inadequate risk management.

[1460] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1461] In this invention, the server includes a means for receiving business content, a means for identifying relevant laws and regulations and potential risks from the received business content, and a generative AI model means for generating countermeasures to address the identified risks. This allows even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters. Furthermore, by including a means for automatically generating necessary documents and emails based on the generated countermeasures, a means for automatically sending the generated documents and emails to relevant parties, and a means for users to input feedback on the content of the automatically generated documents and emails and reflect this feedback in improving the accuracy of the generative AI model, it is possible to further improve business efficiency and the accuracy of risk management.

[1462] "Business content" refers to text information related to a business that a user inputs into a terminal.

[1463] "Terminal means" refers to a device or its interface that allows a user to input business details.

[1464] The "server means" is a server that receives the business contents sent from the terminal and performs the processing.

[1465] "Natural language processing means" refers to technologies and algorithms used to analyze received business content and identify relevant laws and regulations and potential risks.

[1466] "Generative AI model means" refers to an artificial intelligence model for generating countermeasures to identified risks.

[1467] The "automatic document generation means" is a system or program for automatically creating the necessary documents or emails based on the generated solutions.

[1468] "Automatic sending means" refers to a system or method for automatically sending generated documents or emails to relevant parties.

[1469] "Feedback input means" refers to a system or method that allows users to input feedback on the contents of automatically generated documents or emails, and uses that feedback to improve the accuracy of the generative AI model.

[1470] "Template selector" means a system or algorithm for selecting document or email templates based on identified risks and countermeasures.

[1471] The system of the present invention is designed to enable even employees who are not familiar with legal affairs to respond to legal matters promptly and appropriately. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described with reference to specific examples.

[1472] System Configuration

[1473] The system includes the following major hardware and software elements:

[1474] Terminal means: A device for inputting work details. For example, a PC or tablet. The user inputs the work details and sends them to the server.

[1475] Server means: A server that receives and processes the business content sent from the terminal. Specifically, it runs natural language processing (NLP) libraries (e.g., SpaCy, BERT) and generative AI models (e.g., GPT-3).

[1476] Natural language processing means: A program that runs on the server and analyzes the received business content to identify relevant laws and regulations and potential risks.

[1477] Generative AI model means: An artificial intelligence model integrated into the server generates countermeasures for identified risks.

[1478] Automatic document generation: Automatically create documents and emails based on the solutions generated on the server. A template engine (e.g., Jinja2) is used.

[1479] Automatic sending means: A system that sends generated documents and emails to the relevant parties, for example, via an SMTP server.

[1480] Feedback input method: The user checks the contents of the generated documents and emails, and inputs feedback as necessary to improve the accuracy of the generative AI model. The feedback data is stored in a database (e.g., PostgreSQL, MySQL).

[1481] Template selection tools: Tools for selecting document and email templates based on identified risks and treatments.

[1482] Example of a system

[1483] Terminal means

[1484] The user inputs the task details in text format using the input form on the terminal. For example, the task details could be "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[1485] Server Means

[1486] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[1487] Natural language processing and generative AI modeling tools

[1488] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[1489] Automatic document generation method

[1490] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[1491] Automatic transmission method

[1492] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[1493] Feedback Input Method

[1494] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[1495] In this way, the present invention provides a system that enables even employees with insufficient legal knowledge to respond quickly and appropriately to legal matters, and is expected to contribute to improving business efficiency and risk management.

[1496] Specific examples

[1497] For example, if a user inputs "Plan a marketing strategy for new product A," the process will go through the following steps:

[1498] 1. The business details entered by the user are sent to the server from the terminal.

[1499] 2. The server receives the business details and performs natural language processing to identify relevant laws and potential risks. For example, "consumer protection laws" and "regulations on advertising display."

[1500] 3. The server uses the generative AI model to generate a solution, such as suggesting "create advertising copy that complies with the Consumer Protection Act."

[1501] 4. The server automatically generates documents and emails based on the measures taken, such as a "marketing plan" or a "compliance checklist."

[1502] 5. The server automatically sends the generated document to the relevant parties.

[1503] 6. The user reviews the document and provides feedback if necessary.

[1504] Examples of prompts

[1505] An example of a prompt to input to a generative AI model is: "Analyze the laws and potential risks related to the marketing strategy for new product A, and propose appropriate countermeasures."

[1506] Based on this prompt, the generative AI model suggests appropriate actions, and the system then generates the necessary documents and emails.

[1507] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1508] Step 1:

[1509] The user inputs the details of the job into the terminal and sends it to the server.

[1510] Input: Enter the job description in text format as "Plan a marketing strategy for new product A."

[1511] Output: Text data of the work content sent from the terminal to the server.

[1512] Specific operation: The user enters "Plan a marketing strategy for new product A" into the input form on the terminal and clicks the submit button.

[1513] Step 2:

[1514] The server receives the business content and analyzes it using natural language processing means.

[1515] Input: Text data of the work content sent from the terminal.

[1516] Output: Data on relevant laws and regulations and potential risks extracted from business operations.

[1517] Specific operation: The server analyzes the text received from the device, "Plan a marketing strategy for new product A," using a natural language processing tool (such as SpaCy or BERT), and identifies relevant laws and regulations such as "consumer protection laws" and "regulations on advertising display," as well as potential risks.

[1518] Step 3:

[1519] The server uses the generative AI model to generate countermeasures for the identified risks.

[1520] Input: Data on relevant laws and regulations and potential risks identified through natural language processing.

[1521] Output: Text data of relevant laws and regulations and countermeasures for potential risks.

[1522] Specific operation: When a risk related to the Consumer Protection Act is identified, the server sends a prompt to the generative AI model (e.g., GPT-3) saying, "Please create advertising text that complies with the Consumer Protection Act," and generates a solution (appropriate advertising text and display method).

[1523] Step 4:

[1524] The server automatically generates the necessary documents and emails based on the generated solutions.

[1525] Input: Text data of solutions obtained from the generative AI model.

[1526] Output: Auto-generated documents and email data.

[1527] Specific operation: The server receives the generated solutions in text format and uses a template engine (e.g., Jinja2) to automatically generate a "marketing plan in accordance with consumer protection laws" and a "compliance checklist."

[1528] Step 5:

[1529] The server automatically sends generated documents and emails to the relevant parties.

[1530] Input: Data from automatically generated documents and emails.

[1531] Output: The sent document or email reaches the relevant person.

[1532] Specific operation: The server sends the generated "marketing plan" via email to the user, marketing department, and legal department via the SMTP server.

[1533] Step 6:

[1534] The user checks the generated documents and emails and provides feedback.

[1535] Input: The contents of the document or email sent.

[1536] Output: Feedback data from users.

[1537] Specific actions: The user reviews the received "Marketing Plan" and submits feedback by entering a comment such as "This ad copy is appropriate, but I would like it to be a little more specific."

[1538] Step 7:

[1539] The server receives user feedback and uses it to improve the accuracy of the generative AI model.

[1540] Input: Feedback data submitted by the user.

[1541] Output: Updated generative AI model data.

[1542] Specific operation: The server stores the received feedback in a database and uses the feedback data as training data for the generative AI model to improve the accuracy of the model.

[1543] (Application example 1)

[1544] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1545] Conventional regulatory compliance and risk management systems have had difficulty identifying work content and providing appropriate countermeasures quickly and accurately. In particular, with regard to work robots in factories, immediate responses to complex regulations and risks are required, but if on-site personnel lack sufficient legal knowledge, responses may be delayed. Therefore, there is a need for systems that improve the efficiency of regulatory compliance and risk management, thereby improving productivity and safety.

[1546] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1547] In this invention, the server includes a generation AI model means for providing appropriate countermeasures related to received laws and regulations and risks, a terminal means for inputting business content, a server means for receiving the input business content, and a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, thereby enabling prompt and appropriate responses to laws and regulations and risks.

[1548] The "terminal means for inputting work content" refers to a device that allows workers and managers in a factory to input work content and work procedures in voice or text format.

[1549] The "server means for receiving input business content" is a central processing unit for receiving data from the terminal into which the business content has been input, and for managing and processing the data.

[1550] "Natural language processing means for identifying relevant laws and potential risks from received business content" refers to a system that includes natural language processing (NLP) technology for analyzing the text data of received business content and identifying relevant laws and risks.

[1551] "Generative AI model means for generating countermeasures for identified risks" refers to an artificial intelligence (AI) model for generating appropriate countermeasures and guidelines for identified laws and regulations and risks.

[1552] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[1553] "Automatic sending means for automatically sending generated documents and emails to relevant parties" refers to a device or system for automatically sending generated documents and emails to appropriate relevant parties (e.g., factory managers or legal departments).

[1554] "Generative AI model means for providing appropriate countermeasures related to received laws and regulations and risks" is a generative AI model for providing preventive measures and countermeasures based on received laws and regulations and risks.

[1555] The "template selection means" is a system that has the function of selecting an appropriate template according to the identified risks and countermeasures.

[1556] A "feedback input means" is a device or system that allows a user to input feedback on the content of automatically generated documents or emails, and that allows that feedback to be reflected in improving the generative AI model.

[1557] The present invention provides a system for automating a series of processes for compliance with regulations and risk management, from identifying the work content of work robots in a factory to presenting appropriate countermeasures, automatically generating and sending documents, and collecting feedback. A detailed embodiment of this system will be described.

[1558] System configuration

[1559] The system consists of the following hardware and software means:

[1560] 1. Terminal means: A device used to input work content. Input can be done by voice or text. A specific example is the microphone and keyboard installed on a work robot in a factory.

[1561] 2. Server means: A central processing unit that receives and processes input business content. It has a built-in database and analytical model, and performs high-speed and accurate processing.

[1562] 3. Natural Language Processing (NLP): Analyzes business content and identifies relevant laws and regulations and potential risks. For example, a natural language processing library such as Spacy is used.

[1563] 4. Generative AI model means: Generate appropriate countermeasures to address identified risks. For example, the GPT-3 model using the OpenAI API.

[1564] 5. Automatic document generation means: This has the function of automatically creating the necessary documents and emails based on the generated countermeasures. A template-based document generation system is used.

[1565] 6. Automated Delivery: Automatically send generated documents and emails to relevant parties. This includes email systems and internal notification systems.

[1566] 7. Feedback input means: A system in which users can input feedback on the contents of automatically generated documents and emails, and that feedback is reflected in improving the generative AI model.

[1567] Program processing explanation

[1568] The server receives the business content from the terminal and analyzes the input content using natural language processing means. From the analyzed content, relevant laws and regulations and potential risks are identified, and based on that, a generative AI model provides countermeasures. Next, an automatic document generation means automatically generates documents and emails based on the countermeasures, which are then sent to relevant parties using an automatic sending means. The entire process aims to improve business efficiency and ensure swift and appropriate compliance with laws and regulations and risk management.

[1569] Examples of hardware and software used include:

[1570] Hardware: Factory robots, servers, smart devices for administrators

[1571] Software: Spacy (natural language processing library), OpenAI GPT-3 API (generative AI model), email system, template-based document generation system

[1572] Adding specific examples

[1573] For example, when introducing a new product assembly line in a factory, a manager inputs "Introducing a new product assembly line" into the robot's voice input system. The server receives this information and uses natural language processing to identify relevant laws and regulations, such as the Industrial Safety and Health Act, and potential risks, such as "Worker Safety Risks" and "Environmental Pollution Risks." The generative AI model then generates countermeasures based on the following prompt:

[1574] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[1575] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[1576] Please tell me the appropriate countermeasures.

[1577] Based on the proposed measures, the server automatically generates documents such as checklists and guidelines and sends them to the factory manager and legal department. The manager reviews the received documents and provides feedback as necessary, which the system uses to improve the system in the future.

[1578] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[1579] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1580] Step 1:

[1581] The user inputs the job details. A worker or manager in the factory inputs the specific job details (for example, "Introduce an assembly line for a new product") into the work robot's terminal using voice or text format. The input data is saved on the terminal and sent to the server.

[1582] Input: Task details (e.g., "Install an assembly line for a new product")

[1583] Output: Text data transmitted to the terminal, data sent to the server

[1584] Step 2:

[1585] The server receives the work content. The entered work content data is received and prepared for analysis. The received work content is saved to proceed to the next analysis stage.

[1586] Input: Text data sent from the terminal

[1587] Output: Data ready for analysis

[1588] Step 3:

[1589] The business content is analyzed using natural language processing. The server analyzes the received business content using a natural language processing library such as Spacy, and identifies relevant laws and potential risks. For example, relevant laws such as the "Occupational Safety and Health Act" and "Environmental Protection Act" and risks such as "worker safety risks" and "environmental pollution risks" are identified.

[1590] Input: Data ready for analysis

[1591] Output: Identified regulations and potential risks

[1592] Step 4:

[1593] Countermeasures are generated using a generative AI model. The server uses a generative AI model such as OpenAI GPT-3 to generate appropriate countermeasures for the identified laws and potential risks. For this, the following prompt sentences are used:

[1594] Related laws and regulations: Occupational Safety and Health Act, Environmental Protection Act

[1595] Potential risks: "Risk to worker safety", "Risk to environmental pollution"

[1596] Please tell me the appropriate countermeasures.

[1597] Input: Identified laws and regulations and potential risks, prompt text

[1598] Output: Generated countermeasures

[1599] Step 5:

[1600] Documents and emails are automatically generated using an automatic document generation means. Based on the generated countermeasures, the server uses a template-based automatic document generation system to create the necessary documents and emails. For example, "checklists" and "guidelines" are created.

[1601] Input: Generated countermeasures

[1602] Output: Generated documents and emails

[1603] Step 6:

[1604] Send documents and emails to relevant parties using automated methods. The server automatically sends generated documents and emails to relevant parties, such as factory management or the legal department. This process uses email systems and internal notification systems.

[1605] Input: Generated documents and emails

[1606] Output: Documents and emails sent to stakeholders

[1607] Step 7:

[1608] The user enters feedback. Administrators and other relevant parties review the contents of received documents and emails and enter feedback as necessary. This feedback is sent to the server and used to improve the accuracy of the generative AI model.

[1609] Input: User feedback

[1610] Output: Feedback data stored on the server

[1611] In this way, the invention provides a system that enables even factory employees with insufficient legal knowledge to quickly and appropriately comply with regulations and manage risks.

[1612] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1613] The system of the present invention is designed to enable even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and by adding a function to recognize and respond to the user's emotions, the content of documents and emails can be optimized. The following describes in detail the embodiments of the present invention.

[1614] System Overview

[1615] The system mainly consists of the following elements:

[1616] Terminal means for inputting business details

[1617] Server means for receiving input business details

[1618] Natural language processing means to identify relevant laws and potential risks from received business content

[1619] A generative AI model that generates countermeasures to address identified risks

[1620] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[1621] Automatic sending method for automatically sending generated documents and emails to relevant parties

[1622] A means of inputting feedback to contribute to improving the accuracy of generative AI models

[1623] An emotion engine that recognizes emotions based on user input and feedback

[1624] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[1625] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[1626] Example of a system

[1627] Terminal means

[1628] The user inputs the details of the task in text format using the input form on the terminal. For example, the user may input "Plan a marketing strategy for new product A." The terminal means receives this and transmits it to the server means.

[1629] Server Means

[1630] The server receives the business details sent from the terminal. The received information is analyzed using natural language processing. As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, if the business details are related to marketing strategies, relevant laws and regulations such as consumer protection laws and advertising regulations are identified.

[1631] Natural language processing and generative AI modeling tools

[1632] The server analyzes the received business content and identifies relevant laws and potential risks. It then uses a generative AI model to generate countermeasures for the identified risks. For example, if a risk of violating advertising regulations is identified, the server will suggest countermeasures such as selecting appropriate advertising wording and displaying methods that comply with laws and regulations.

[1633] Automatic document generation method

[1634] The server automatically generates the necessary documents and emails based on the generated solutions. At this time, it selects templates according to the business content and solutions and creates specific documents and emails. For example, it automatically generates marketing plans and compliance checklists based on the Consumer Protection Act.

[1635] Automatic transmission method

[1636] The generated documents and emails are automatically sent to the appropriate parties, including the user, relevant internal departments (marketing, legal, etc.), and external legal counsel.

[1637] Feedback Input Method

[1638] Users can review the content of automatically generated documents and emails and provide feedback if necessary, which is sent to the server and used to improve the accuracy of the generative AI model.

[1639] Emotion engine and wording optimization

[1640] The emotion engine analyzes user input and feedback and recognizes the emotion. For example, if a user inputs an urgent task, the emotion engine recognizes this as the emotion representing "urgent." The recognized emotion is used by the wording optimization tool to optimize the tone and wording of generated documents and emails. This allows recipients to better understand the content and respond quickly.

[1641] Emotion data reflection method

[1642] The emotion data recognized by the emotion engine is also reflected in the learning data of the generative AI model, improving the model's accuracy and flexibility, and enabling more appropriate responses in future business content analysis and document generation.

[1643] Specific examples

[1644] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[1645] 1. The user enters the details of the job into the terminal and sends it to the server.

[1646] 2. The server analyzes the business content and identifies relevant laws and regulations such as consumer protection laws and advertising display regulations.

[1647] 3. The server requests the generative AI model to generate risk countermeasures and receives the appropriate countermeasures.

[1648] 4. The server automatically generates documents and emails based on the corrective action.

[1649] 5. Language optimization tools adjust the tone of documents and emails based on user sentiment.

[1650] 6. The server automatically sends the generated document to the marketing and legal departments.

[1651] 7. The user checks the results on the device and provides feedback if necessary.

[1652] 8. The feedback content is sent to the server, where it is emotionally analyzed by the emotion engine and reflected in the training data of the generative AI model.

[1653] In this way, the present invention is highly effective in improving business efficiency and risk management by automating the generation and transmission of legal documents that reflect the user's feelings.

[1654] The processing flow will be explained below.

[1655] Step 1:

[1656] The user enters the details of the task in text format into the input form on the terminal. For example, the user enters "Plan the marketing strategy for new product A."

[1657] Step 2:

[1658] The device sends the entered business details to the server using an API, which transfers the data securely.

[1659] Step 3:

[1660] The server receives the submitted business details and passes the received data to the natural language processing engine.

[1661] Step 4:

[1662] The server uses a natural language processing engine to analyze the received business content. As a result of the analysis, the structure and meaning of the sentence are understood, and relevant laws and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" are identified.

[1663] Step 5:

[1664] The server requests the generative AI model to generate a solution to the identified risk. The generative AI model then references relevant databases and past cases to generate the optimal solution.

[1665] Step 6:

[1666] The generative AI model responds to the server with suggestions for how to deal with the problem, such as selecting appropriate advertising copy or displaying the ad in a way that complies with regulations.

[1667] Step 7:

[1668] The server selects a document or email template based on the proposed solution. Select a template that suits the business content and solution.

[1669] Step 8:

[1670] The server uses a document generation engine to automatically generate necessary documents and emails, such as marketing plans and compliance checklists based on consumer protection laws.

[1671] Step 9:

[1672] The user enters feedback from the terminal, providing additional information and suggestions for improvement based on the content of the generated documents and emails.

[1673] Step 10:

[1674] The server receives the feedback and passes it to the emotion engine for analysis. The user's emotion is identified from the feedback content. For example, the emotion meaning "urgency" is recognized.

[1675] Step 11:

[1676] The server optimizes the wording of documents and emails based on the emotional data from the emotion engine, adjusting the tone and wording as needed.

[1677] Step 12:

[1678] The server automatically sends the generated documents and emails to various parties, including the marketing department, legal department, and external legal counsel.

[1679] Step 13:

[1680] The server reflects the emotion data identified by the emotion engine in the learning data of the generative AI model, improving the accuracy of the generative AI model and making future analysis and generation more accurate.

[1681] In this way, the system automates a series of processes, from inputting work details to analyzing relevant laws and risks, generating countermeasures, automatically generating and sending documents, improving accuracy through feedback, and even optimizing wording by combining an emotion engine. This process enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal issues, improving work efficiency and risk management.

[1682] Example 2

[1683] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1684] There is a demand for a system that can respond to legal issues quickly and appropriately, even when employees are not familiar with legal matters. It is also necessary to optimize the content of documents and emails by reflecting the user's feelings, and ensure smooth correspondence between the parties involved. Conventional systems do not automate the identification of laws and regulations or risk countermeasures, which means that legal responses take time, and the system does not reflect the user's feelings, making it difficult to communicate appropriately.

[1685] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1686] In this invention, the server includes a terminal means for inputting business content, a server means for receiving the input business content, a natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, a generative AI model means for generating countermeasures to address the identified risks, an automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, an automatic sending means for automatically sending the generated documents and emails to relevant parties, a feedback input means for allowing a user to input feedback on the content of the automatically generated documents and emails and reflecting this feedback in improving the accuracy of the generative AI model, an emotion engine for recognizing a user's emotions based on the user's input content and feedback, a wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine, and an emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means. This enables even employees who are not familiar with legal matters to respond quickly and appropriately to legal matters, and documents and emails that reflect the user's emotions can be generated and sent, thereby facilitating communication.

[1687] The "terminal means for inputting business content" is a device equipped with an input device and an input form for a user to input business content in text format.

[1688] The "server means for receiving the input business content" is a server device for receiving the business content input by the user through the terminal means.

[1689] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to means that uses natural language processing technology to analyze input business content and identify relevant laws and regulations and potential risks.

[1690] A "generative AI model means for generating countermeasures for identified risks" is a means for using an artificial intelligence model to generate appropriate countermeasures for identified risks.

[1691] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a means for automatically creating necessary documents and e-mails based on the generated solutions.

[1692] The "automatic sending means for automatically sending the generated document or e-mail to the relevant person" is a means for automatically sending the generated document or e-mail to the designated relevant person.

[1693] "Feedback input means that allows users to input feedback on the content of automatically generated documents or emails and reflect it in improving the accuracy of the generative AI model" refers to a means that allows users to input opinions and suggestions for improvement on the content of automatically generated documents or emails, and uses that feedback information as learning data for the generative AI model.

[1694] "Emotion engine that recognizes user emotions based on user input and feedback" is an engine that analyzes and recognizes emotions from the content and feedback entered by the user.

[1695] The "wording optimization means for optimizing the wording of documents and emails based on the user's emotions recognized by the emotion engine" is a means for optimally adjusting the content of generated documents and emails based on the user's emotions recognized by the emotion engine.

[1696] The "emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generative AI model means" is a means for incorporating the emotion data recognized by the emotion engine into the learning data of the generative AI model.

[1697] The "template selection means" is a means for selecting an appropriate template for a document or email based on the identified risks and countermeasures.

[1698] The system of the present invention is designed to process legal work quickly and appropriately, and enables the generation of documents and emails that reflect the user's feelings. A specific embodiment of the present invention will be described below.

[1699] 1. System Configuration

[1700] The system consists of the following elements:

[1701] Terminal means for inputting business details

[1702] Server means for receiving input business details

[1703] Natural language processing means to identify relevant laws and potential risks from received business content

[1704] A generative AI model that generates countermeasures to address identified risks

[1705] An automatic document generation method that automatically generates necessary documents and emails based on the generated solutions

[1706] Automatic sending method for automatically sending generated documents and emails to relevant parties

[1707] A feedback input method that allows users to input feedback on the content of automatically generated documents and emails, and reflects this feedback in improving the accuracy of the generative AI model.

[1708] An emotion engine that recognizes user emotions based on user input and feedback

[1709] A wording optimization method that optimizes the wording of documents and emails based on the user's emotions recognized by an emotion engine

[1710] Emotion data reflection means for reflecting the emotion data recognized by the emotion engine in the learning data of the generation AI model means

[1711] 2. Hardware and Software Used

[1712] Terminal means: The user uses an input device such as a computer or tablet to enter the details of the work in text format into a dedicated input form.

[1713] Server means: The server is a platform for receiving and processing business content sent from the terminal. The server may also use cloud-based services.

[1714] Natural language processing: For natural language processing, we use NLP libraries such as SpaCy and BERT to analyze the input business content and identify relevant laws and regulations and potential risks.

[1715] Generative AI model means: To generate countermeasures, we use generative AI models such as GPT-3 and BERT, which automatically generate appropriate countermeasures.

[1716] Automatic document generation method: To automatically generate documents and emails, a specified template is used. The template is automatically selected according to the business content and the solution.

[1717] Emotion Engine: For emotion recognition, emotion analysis algorithms (e.g., TextBlob, VADER) are used to identify emotions from user input and feedback.

[1718] 3. Specific Examples

[1719] For example, if a user inputs "Plan a marketing strategy for new product A," the following process will be performed automatically.

[1720] 1. Using the terminal means, the user enters "Plan a marketing strategy for new product A" into the input form and presses the send button.

[1721] 2. The terminal sends the entered business details to the server.

[1722] 3. The server analyzes the received data using natural language processing tools (e.g., SpaCy) to identify consumer protection laws and advertising regulations.

[1723] 4. The server sends a prompt to the generative AI model saying, "Please generate solutions to address the risk of violating advertising regulations," and receives the solutions.

[1724] 5. The server automatically generates a marketing plan using a template based on the solution.

[1725] 6. The server automatically sends the generated plan to the marketing and legal departments.

[1726] 7. The user checks the plan and enters feedback such as "Please correct this part."

[1727] 8. The server analyzes the feedback, recognizes the emotion of "urgency," and adjusts the tone of the document using language optimization techniques.

[1728] 9. The server reflects the recognized emotion data in the learning data of the generative AI model, aiming to improve accuracy from the next time onwards.

[1729] Example prompt sentence:

[1730] "Identify the laws and regulations and potential risks that need to be considered when developing a marketing strategy for new product A, and generate appropriate countermeasures."

[1731] "Generate appropriate documentation to comply with consumer protection laws and advertising regulations when developing your marketing strategies."

[1732] In this way, the present invention enables even users who are not familiar with legal matters to respond to legal matters quickly and appropriately, and facilitates smooth communication through the creation and transmission of documents and emails that reflect emotions.

[1733] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1734] Step 1:

[1735] The user uses the terminal means to input the details of the job in text format into the input form. For example, the user might input "Plan a marketing strategy for new product A" and press the send button.

[1736] Input: Job description (e.g., "Plan a marketing strategy for new product A")

[1737] Output: Text data of the work content is sent from the terminal

[1738] Step 2:

[1739] The terminal transmits the text data of the entered business details to the server.

[1740] Input: Text data of business content

[1741] Output: Data sent to the server

[1742] Step 3:

[1743] The server analyzes the received text data of the business content using natural language processing tools (e.g., SpaCy). As a result of the analysis, relevant laws and regulations and potential risks are identified. For example, "consumer protection laws" and "regulations on advertising display" may be identified.

[1744] Input: Text data of business content

[1745] Data processing: Applying natural language processing to identify relevant laws and risks

[1746] Output: List of relevant laws and regulations and potential risks

[1747] Step 4:

[1748] The server requests the generative AI model to generate a solution based on the identified laws and risks. For example, it sends a prompt saying, "Please generate a solution to address the risk of violating advertising display regulations." The generative AI model generates an appropriate solution and sends it back to the server.

[1749] Input: List of relevant laws and regulations and potential risks, prompt text

[1750] Data computation: Using generative AI models to generate solutions

[1751] Output: A list of appropriate actions

[1752] Step 5:

[1753] Based on the received solutions, the server uses an automatic document generation means to automatically generate the necessary documents and emails. At this time, a template appropriate for the business content and solutions is selected, and specific documents and emails are created. For example, a "marketing plan" or "compliance checklist" is automatically generated.

[1754] Input: Solution list, template

[1755] Data processing: Generate documents and emails based on solutions and templates

[1756] Output: Generated documents and emails

[1757] Step 6:

[1758] The server automatically sends the generated documents and emails to the appropriate parties, including the user, relevant internal departments (e.g., marketing, legal), and external legal counsel.

[1759] Input: Generated documents and emails

[1760] Data processing: Send using email transmission protocols (e.g., SMTP)

[1761] Output: Documents and emails sent to stakeholders

[1762] Step 7:

[1763] The user checks the content of the automatically generated document or email and enters feedback as needed. For example, the user may enter feedback such as "Please correct this part." The feedback is sent from the terminal to the server.

[1764] Input: Feedback

[1765] Output: Feedback is sent to the server

[1766] Step 8:

[1767] The server receives the feedback and uses an emotion engine to analyze the user's emotions. For example, the emotion representing "urgency" is recognized. Based on this emotion, a wording optimizer adjusts the tone and wording of the generated document or email.

[1768] Input: Feedback

[1769] Data Computation: Emotion Analysis with Emotion Engine

[1770] Output: Recognized sentiment, optimized wording

[1771] Step 9:

[1772] The server reflects the emotion data recognized by the emotion engine in the learning data of the generative AI model, enabling more appropriate responses in subsequent analyses and document generation.

[1773] Input: Emotion data

[1774] Data processing: Reflecting emotion data in the training data of the generative AI model

[1775] Output: Updated training data for the generative AI model

[1776] In this way, a system is realized in which the processing at each step is linked, allowing even users unfamiliar with legal matters to respond quickly and appropriately to legal matters.

[1777] (Application example 2)

[1778] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1779] With conventional systems, it was difficult for employees without legal expertise to generate documents that complied with regulations, which increased the likelihood of mistakes and risks. Furthermore, the system was unable to respond flexibly to user sentiment, resulting in reduced operational efficiency. These issues must be resolved, particularly for online shopping sites, where compliance with regulations and communication based on user sentiment are crucial.

[1780] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes server means for receiving input business content, natural language processing means for identifying relevant laws and regulations and potential risks from the received business content, generation AI model means for generating countermeasures for addressing the identified risks, automatic document generation means for automatically generating necessary documents and emails based on the generated countermeasures, automatic sending means for automatically sending the generated documents and emails to relevant parties, emotion analysis means for recognizing emotions from user input content and feedback, and wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means. This enables appropriate responses to laws and regulations and flexible document generation based on user emotions.

[1781] The "terminal means for inputting business details" is a device equipped with an interface for a user to input business details in text format.

[1782] The "server means for receiving input business content" is a server for receiving and storing data sent by a user from a terminal means.

[1783] "Natural language processing means for identifying relevant laws and regulations and potential risks from received business content" refers to natural language processing technology for analyzing received text data and identifying relevant laws and regulations and potential risks.

[1784] A "generative AI model means for generating countermeasures to address identified risks" is an artificial intelligence model that generates methods and guidelines for addressing identified risks.

[1785] The "automatic document generation means for automatically generating necessary documents and e-mails based on the generated solutions" is a system for automatically creating necessary documents and e-mails based on the generated solutions.

[1786] "Automatic sending means for automatically sending generated documents and e-mails to relevant parties" refers to a system that has the function of automatically sending generated documents and e-mails to designated relevant parties.

[1787] "Emotion analysis means for recognizing emotions from user input and feedback" is a technology for determining emotions from text data and feedback entered by the user.

[1788] The "wording optimization means for optimizing the wording of documents and emails based on the emotions recognized by the emotion analysis means" is a system that adjusts the tone and expression of generated documents and emails based on the emotions recognized by the emotion analysis means.

[1789] The "template selection means for selecting a document or email template based on identified risks and countermeasures" is a system that has the function of automatically selecting the most appropriate template based on risks and countermeasures.

[1790] "Feedback input means that allows users to input feedback on the contents of automatically generated documents and emails, and reflect this in improving the accuracy of the generative AI model" refers to a system that has the function of allowing users to provide feedback on automatically generated documents and emails, and to improve the generative AI model based on that feedback.

[1791] "A means for updating a generative AI model that reflects feedback content and emotional data based on generated documents and emails in the learning data of the generative AI model" is a technology that uses provided feedback and emotional data to update a generative AI model and improve the accuracy and flexibility of the model.

[1792] This embodiment of the present invention relates to a compliance assistance system for online shopping sites. The purpose of this system is to quickly and accurately perform legal checks when users release new products or services. A specific embodiment of this system will be described below.

[1793] System configuration

[1794] This system mainly consists of the following hardware and software:

[1795] Hardware: Smartphone

[1796] Software: Python, natural language processing libraries (spaCy, NLTK), sentiment analysis library (TextBlob), generative AI model (OpenAI GPT-3), Flask, REST API, Jinja2, smtplib

[1797] Program processing

[1798] 1. Input form: Using a smartphone app, users enter text descriptions of new products and services. This input form is built using HTML and JavaScript.

[1799] 2. Data reception: The data entered by the user is sent to the server via a REST API built using Flask. The server receives and stores this data.

[1800] 3. Natural Language Analysis: The server analyzes the received input data using Python and natural language processing libraries (spaCy, NLTK). This analysis identifies relevant laws and regulations and potential risks.

[1801] 4. Generative AI model: Based on the identified risks, a generative AI model (GPT-3) is used to generate countermeasures. The generative AI model is invoked using Python and the OpenAI API.

[1802] 5. Sentiment Analysis: TextBlob is used to analyze sentiment from user text input and feedback. This sentiment data is used to generate documents and emails.

[1803] 6. Automatic document generation: Based on the generated responses and the results of sentiment analysis, the necessary documents and emails are automatically generated using Jinja2.

[1804] 7. Automatic sending: Generated documents and emails are automatically sent to the relevant parties using Python's smtplib library.

[1805] 8. Feedback collection: Users can input feedback on the content of generated documents and emails through a smartphone app. This feedback is sent back to the server and reflected in the training data for the generative AI model.

[1806] In this way, appropriate measures for legal regulations and flexible document generation based on user feelings are realized.

[1807] Specific examples

[1808] For example, the user enters the following prompt text:

[1809] Develop a marketing strategy for new product A, including how to effectively appeal to the target market while complying with consumer protection laws and advertising regulations.

[1810] The server that receives this input uses natural language processing means to identify relevant laws and regulations, such as "consumer protection laws" and "regulations on advertising display," as well as risks. Then, the generative AI model means generates countermeasures for these risks. Based on the generated countermeasures, the automatic document generation means creates appropriate marketing plans and compliance checklists.

[1811] The sentiment analysis means analyzes t...

Claims

1. a terminal means for inputting business details; A server means for receiving input business details; natural language processing means for identifying relevant laws and regulations and potential risks from the received business content; a generative AI model means for generating a response to the identified risk; an automatic document generation means for automatically generating necessary documents and emails based on the generated solutions; An automatic sending means for automatically sending the generated documents and emails to the relevant parties; A system including:

2. The system according to claim 1, further comprising a template selection means for selecting a document or email template based on the identified risks and countermeasures.

3. The system according to claim 1, further comprising a feedback input means for allowing a user to input feedback on the content of automatically generated documents or emails, and for reflecting this feedback in improving the accuracy of the generated AI model.

Images